For months I’ve been concerned that I have difficulty telling the difference between good science and bad science in the latest technical articles in astrophysics. Perhaps I’ve go so behind the times that I can no longer understand technical science papers. Or perhaps not, it occurred to me just now that perhaps I’m running into Sturgeon’s Law.

Sturgeon’s law is “90% of everything is crap”. But is it really 90%?
The first step in answering that is to identify what is and isn’t crap when applied to technical science articles.

A) Crap science includes:
1) Papers on ramifications of scientific hypotheses that have already been disproved (including those rejected by Occam’s razor).
2) Observations that are not statistically significant.
3) Hobby horse science – the same paper published over and over again with only trivial modifications.
4) Minimal effort, would have required less than three man-days to research.
5) Just plain wrong.

B) There there are papers that, while not complete crap, do not advance science:
1) Just another paper about @#$%. ie. a hobby horse but on a global scale.
2) Minutiae that add only a trivial amount of new information.
3) Confirming what everyone already knows, to no more accuracy than earlier studies.
4) Timid science, so afraid of offending people that it doesn’t come to any conclusions.
5) An un-approved proposal.

C) Then there’s average science:
1) New observations or theory for something that nobody cares about.
2) New observations that supersede older ones.
3) A new science facility.
4) Confirming a discrepancy between theory and observation.
5) New (unconfirmed) hypothesis that could explain the discrepancy between theory and observation.

D) Then there’s great science:
1) Observed for the first time, and important.
2) New statistically significant discrepancy between theory and observation.
3) A newly-confirmed theory about something important.

E) Other:
1) I haven’t got a clue what they’re talking about.

> B3) Confirming what everyone already knows, to no more accuracy than earlier studies.

This seems to be the most common, some examples I’ve noticed are:

A new calculation of Mercury’s orbital parameters to less accuracy than had been done previously
Extending a calculation based on 5.2 years of Fermi data to one based on 5.8 years of Fermi data.

But what if there’s an instance of:

4) Minimal effort, would have required less than three man-days to research.

…that also turns out to be one of these:

1) Observed for the first time, and important.
2) New statistically significant discrepancy between theory and observation.
3) A newly-confirmed theory about something important.

Bubblecar said:

But what if there’s an instance of:

4) Minimal effort, would have required less than three man-days to research.

…that also turns out to be one of these:

1) Observed for the first time, and important.
2) New statistically significant discrepancy between theory and observation.
3) A newly-confirmed theory about something important.

There have of course been many such instances, especially in regard to 1).

mollwollfumble said:

A) Crap science includes:
1) Papers on ramifications of scientific hypotheses that have already been disproved (including those rejected by Occam’s razor).

Forgive me for jumping in at Point A 1, but isn’t rejecting something on the basis of Occam’s Razor an example of the 90% described by Sturgeon’s Law?

The Rev Dodgson said:

mollwollfumble said:

A) Crap science includes:
1) Papers on ramifications of scientific hypotheses that have already been disproved (including those rejected by Occam’s razor).

Forgive me for jumping in at Point A 1, but isn’t rejecting something on the basis of Occam’s Razor an example of the 90% described by Sturgeon’s Law?

mollwollfumble said:

A) Crap science includes:
3) Hobby horse science – the same paper published over and over again with only trivial modifications.

B) There there are papers that, while not complete crap, do not advance science:
2) Minutiae that add only a trivial amount of new information.
3) Confirming what everyone already knows, to no more accuracy than earlier studies.

C) Then there’s average science:
1) New observations or theory for something that nobody cares about.
2) New observations that supersede older ones.

D) Then there’s great science:

E) Other:
1) I haven’t got a clue what they’re talking about.

As for the rest of them, I’ve listed above the ones I see a lot of (talking mainly engineering science here).

So I wouldn’t say 90% crap, but if we made it crap to pretty mediocre, then probably 95%.

Still, I suppose 5% really good stuff isn’t bad.

The Rev Dodgson said:

Still, I suppose 5% really good stuff isn’t bad.

When you look at the IQ curve…

Is 90% of Sturgeon’s Law crap?

The Rev Dodgson said:

Is 90% of Sturgeon’s Law crap?

No. Sounds more like, fishy.

roughbarked said:

The Rev Dodgson said:

Is 90% of Sturgeon’s Law crap?

No. Sounds more like, fishy.

Depends how you carve it up I suppose.

The Rev Dodgson said:

roughbarked said:

The Rev Dodgson said:

Is 90% of Sturgeon’s Law crap?

No. Sounds more like, fishy.

Depends how you carve it up I suppose.

Could get some exceptional fillets.

It’s the fish that Occam’s Law reject, that makes Occam’s Law contentious.

Bubblecar said:

It’s the fish that Occam’s Law reject, that makes Occam’s Law contentious.

Actually “razor”, it’s not usually enshrined as a Law :)

Bubblecar said:

Bubblecar said:

It’s the fish that Occam’s Law reject, that makes Occam’s Law contentious.

Actually “razor”, it’s not usually enshrined as a Law :)

I suppose Occam’s Razor would be the implement of choice of a Sturgeon Surgeon.

I’m nonetheless sympathetic to Occam’s Razor and can appreciate its work when wielded judiciously.

The Rev Dodgson said:

Bubblecar said:

Bubblecar said:

It’s the fish that Occam’s Law reject, that makes Occam’s Law contentious.

Actually “razor”, it’s not usually enshrined as a Law :)

I suppose Occam’s Razor would be the implement of choice of a Sturgeon Surgeon.

Caviar comes from the virgin Sturgeon,
Virgin Sturgeon might fine fish,
Virgin Sturgeon needs no uregin
That’s why caviar is my dish.

bob(from black rock) said:

The Rev Dodgson said:

Bubblecar said:

Actually “razor”, it’s not usually enshrined as a Law :)

I suppose Occam’s Razor would be the implement of choice of a Sturgeon Surgeon.

Caviar comes from the virgin Sturgeon,
Virgin Sturgeon might fine fish,
Virgin Sturgeon needs no uregin
That’s why caviar is my dish.

That’s beautiful.

The Rev Dodgson said:

roughbarked said:

The Rev Dodgson said:

Is 90% of Sturgeon’s Law crap?

No. Sounds more like, fishy.

Depends how you carve it up I suppose.

Sturgeon’s Law crap?
Do you mean Carp? not crap?

> But what if there’s an instance of …that also turns out to be one of these:

I allocate it to the most valuable category.

> Isn’t rejecting something on the basis of Occam’s Razor an example of the 90% described by Sturgeon’s Law?

That’s what I want to find out. Take for example the f® or Brans-Dicke theories of gravity. These can never be ruled out without Occam’s Razor because the both contain General Relativity as a special case. But neither can the Hooloovoo (a hyper-intelligent shade of the colour blue) be ruled out without Occam’s Razor. I’ll accept NEW gravity theories that contain General Relativity as a special case as non-crap, but not the repeat of old theories.

> As for the rest of them, I’ve listed above the ones I see a lot of

Preliminary results here, I totally agree with you. So far, B3 is most common followed by C1 then C2, then eight of the others.

I was initially aiming to evaluate results based on 1000 papers from arxiv/astro-ph from June. But I haven’t even got through the first 50 yet. Already I’ve had to add one more category:
C6) A mathematical method that isn’t very new.

I shudder to quote here my example so far of A5) just plain wrong.
It’s a paper that claims that the spacing of the Moons of Uranus is due to the spacing of spectral lines of hydrogen.

My example of B2) so far is a calculation of the spectral lines of NaH for astronomical use.
OK you might think, but the paper concludes that no NaH has ever been observed in any astronomical spectrum.

mollwollfumble said:

> But what if there’s an instance of …that also turns out to be one of these:

I allocate it to the most valuable category.

> Isn’t rejecting something on the basis of Occam’s Razor an example of the 90% described by Sturgeon’s Law?

That’s what I want to find out. Take for example the f® or Brans-Dicke theories of gravity. These can never be ruled out without Occam’s Razor because the both contain General Relativity as a special case. But neither can the Hooloovoo (a hyper-intelligent shade of the colour blue) be ruled out without Occam’s Razor. I’ll accept NEW gravity theories that contain General Relativity as a special case as non-crap, but not the repeat of old theories.

> As for the rest of them, I’ve listed above the ones I see a lot of

Preliminary results here, I totally agree with you. So far, B3 is most common followed by C1 then C2, then eight of the others.

I was initially aiming to evaluate results based on 1000 papers from arxiv/astro-ph from June. But I haven’t even got through the first 50 yet. Already I’ve had to add one more category:
C6) A mathematical method that isn’t very new.

I shudder to quote here my example so far of A5) just plain wrong.
It’s a paper that claims that the spacing of the Moons of Uranus is due to the spacing of spectral lines of hydrogen.

My example of B2) so far is a calculation of the spectral lines of NaH for astronomical use.
OK you might think, but the paper concludes that no NaH has ever been observed in any astronomical spectrum.

> the f®
Darn autocorrect.
That’s the eff-open-bracket-r-close-bracket theory of gravity.

I’m beginning to wonder if some of the large number of papers that appear to be of negligible importance are actually deliberately underselling themselves. If they sell themselves as “this is a minor extension of previous work” then they are more likely to get a favourable peer review than if they sell themselves as “this is a radical new development”.

ie. Scientific papers probably undersell themselves in order to satisfy “The Peter Principle”. ie. Peer review leads to: “Rather than seeking to promote a talented super-competent junior peer, an incompetent senior peer will reject them because they are likely to violate the first commandment of hierarchical life with incompetent leadership: namely that the hierarchy must be preserved”.

mollwollfumble said:

I shudder to quote here my example so far of A5) just plain wrong.
It’s a paper that claims that the spacing of the Moons of Uranus is due to the spacing of spectral lines of hydrogen.

Hmmm, maybe I should review my engineering papers with a more critical eye.

I bet I could find something equally crazy with a bit of effort.

mollwollfumble said:

I’m beginning to wonder if some of the large number of papers that appear to be of negligible importance are actually deliberately underselling themselves. If they sell themselves as “this is a minor extension of previous work” then they are more likely to get a favourable peer review than if they sell themselves as “this is a radical new development”.

ie. Scientific papers probably undersell themselves in order to satisfy “The Peter Principle”. ie. Peer review leads to: “Rather than seeking to promote a talented super-competent junior peer, an incompetent senior peer will reject them because they are likely to violate the first commandment of hierarchical life with incompetent leadership: namely that the hierarchy must be preserved”.

yes but in practice “networking” takes over – its not what you know and what you can do – its who you know, its why richer parents send there children to private expensive schools – you network with the people there and with the students with other privates schools, when you need a job or need to be slid into a good job you’ll most likely hear it from someone who knows someone.

the other thing is this – when someone promotes you their worth is at risk. your incompetence can’t be exposed because it reflects badly on the competence of the person who put you into the role in the first place

management is a hive mentality because any question of competence ultimately leads to all levels of management within the group,

hence my assertion : “management can never be wrong”

B) 3) Confirming what everyone already knows, to no more accuracy than earlier studies

The older I get, the more of these I see in my area of expertise. I personally refer to them as “Well, derrrr!!” papers. Which probably shows I am a pre Simpson’s person. I do think researchers should read what has gone before.

mollwollfumble said:

I’m beginning to wonder if some of the large number of papers that appear to be of negligible importance are actually deliberately underselling themselves. If they sell themselves as “this is a minor extension of previous work” then they are more likely to get a favourable peer review than if they sell themselves as “this is a radical new development”.

ie. Scientific papers probably undersell themselves in order to satisfy “The Peter Principle”. ie. Peer review leads to: “Rather than seeking to promote a talented super-competent junior peer, an incompetent senior peer will reject them because they are likely to violate the first commandment of hierarchical life with incompetent leadership: namely that the hierarchy must be preserved”.

I think you really might be onto something there.

once a management structure is cemented into place its a race to the bottom

when things go badly if you are a manager and are responsible for the problem you quickly move on to destroy somewhere else with the same practices that destroyed your last lace of employment

when I worked at one place chemicals were stored in large wooden lockers that doubled as storage laces for our personal effects, on the first day I saw various chemicals of different colours leaking out the front of the lockers

even more alarming was that asbestos removal was being done with no normal safeguards – I was the only person in the place wit any from of training including the “safety officer”

when I demonstrated that high quality stainless steel could only be drilled through using a special type of drill bit it was rejected , they eventually bought a small drill set but it was stolen. they told me they didn’t have enough money for cobalt drill bits for SS – yet the company gave 1 million dollars to “art”

Australian industry suffers from the management effect – its why it wont much longer

Is it a publish or perish kind of thing?

Dropbear said:

Is it a publish or perish kind of thing?

I suspect so. Where are our academic scientists these days?

> A5) Just plain wrong
> D2) New statistically significant discrepancy between theory and observation.

Not sure whether this paper qualifies as D2) or A5). There appears to be no correlation between the location of gas in star-forming regions and the location of stars in star-forming regions.

As a thought experiment “high gas density” → “more stars formed” but the radiation from newly formed stars drives away the gas so “more stars formed” → “low gas density”. So perhaps they’re combining star forming regions of different ages, in younger ones there’s a positive correlation and older ones there’s a negative correlation. Yes they are, so it’s A5).

As you were.

Here’s a weird one. Combining ages and directions of three nearby star-forming regions – in Taurus, in Ophiuchus and in the Gould Belt – it seems possible that all three were in the vicinity of our solar system 24 to 26 million years ago. Not necessarily close vicinity, but 20 to 30 times as close as Taurus and Ophiuchus are now. This appears to be a new observation, and could be important.

When GAIA has finished its observations in four or so years we’ll know a lot more.

Results from 100 arXiv (first hundred from June 2015). A hundred is nowhere near enough for a reliable analysis of this sort, because the timing of articles placed in arXiv is far from random. Perhaps best to add a +-5%.

mollwollfumble said:

For months I’ve been concerned that I have difficulty telling the difference between good science and bad science in the latest technical articles in astrophysics. Perhaps I’ve go so behind the times that I can no longer understand technical science papers. Or perhaps not, it occurred to me just now that perhaps I’m running into Sturgeon’s Law.

Sturgeon’s law is “90% of everything is crap”. But is it really 90%?

Results.

A) Crap science (13%)
5% 1) Papers on ramifications of scientific hypotheses that have already been disproved (including those rejected by Occam’s razor).
1% 2) Observations that are not statistically significant.
3% 3) Hobby horse science – the same paper published over and over again with only trivial modifications.
2% 4) Minimal effort, would have required less than three man-days to research.
2% 5) Just plain wrong.

B) Papers that, while not complete crap, do not advance science much (55%)
3% 1) Just another paper about @#$%. ie. a hobby horse but on a global scale.
1% 2) Minutiae that add only a trivial amount of new information.
35% 3) Confirming what is already known (or confirming that nobody knows) to no more accuracy than earlier studies.
0% 4) Timid science, so afraid of offending people that it doesn’t come to any conclusions.
4% 5) An un-approved proposal.
12% 6) A mathematical method that isn’t very new.

C) There’s average science (30%)
13% 1) New observations or mathematical modelling of something that nobody cares about.
12% 2) New observations that supersede older ones.
0% 3) A new science facility.
1% 4) Confirming a discrepancy between theory and observation.
4% 5) New (unconfirmed) hypothesis or new observations that could reduce the discrepancy between theory and observation.

D) There’s great science (2%)
2% 1) Observed for the first time, and important.
0% 2) New statistically significant discrepancy between theory and observation.
0% 3) A newly-confirmed theory about something important.

E) Other:
0% 1) I haven’t got a clue what they’re talking about.

In summary, to apply Sturgeon’s law to scientific articles in hard science, you need to replace that “90%” by “about 70%”.

Some of the more “interesting” articles.
1% A2) Observations that are not statistically significant.
Observations of TiO2 in the atmosphere of a giant star

3% B1) Just another paper about @#$%. ie. a hobby horse but on a global scale.
I expected heaps of “black hole” papers here, but they mostly avoided me this time. The first of the two I did count as global hobby horses in this dataset is the current obsession with variability of active galactic nuclei. It’s been known for at least 40 years that this is intrinsically unpredictable, as all the latest papers confirm. The second is theories of cosmic inflation, anything not known about cosmic inflation from the results of the Planck space telescope will remain unknown for a very long time. Not counted as a global hobby horse but surprisingly popular is the magnetic field in intergalactic space, I gather this was recently discovered to be much stronger than expected, but now everyone agrees why.

B3) Confirming that nobody knows.
The most startling example of this is a paper confirming that we haven’t a clue how much oxygen is in the Sun. Spectral line evidence is grossly inconsistent, and even more inconsistent unless we have the Sun’s nickel content wrong, and helio-seismicity doesn’t clarify the issue.

0% B4) Timid science.
I had that on the list because that’s a description of the papers describing the final cosmological results from the Planck satellite.

12% C2) New observations that supersede older ones.
I have two favourites here. In one, combining data from SDSS and WISE found that our previous estimation of the star formation rate in the universe was too high by a factor of 1.4. The other is a 15 month result from Pan-STARRS, which includes absolute magnitudes of 250,000 asteroids.

1% C4) Confirming a discrepancy between theory and observation.
The atmosphere of Pluto is denser, contains more nitrogen, than it should. Keep a watch on the results from New Horizons to see if Pluto has any nitrogen volcanos.

2% D1) Observed for the first time, and important.
Supernovae type 1a are the “standard candles” of the cosmos. For the first time we have a time-series of spectrographs from a nearby example, in the Cigar galaxy M82.

0% E1) I haven’t got a clue what they’re talking about.
(Pat self on back).

Category A4 seems harsh. You can do useful science in less than 2 days.

Also we need D4:

Stuff we thought was just plain wrong, but it isn’t.

The Rev Dodgson said:

Also we need D4:

Stuff we thought was just plain wrong, but it isn’t.

Then there’s stuff we thought was wrong, then wasn’t wrong, then was wrong again.

Bubblecar & Dodgson said:

Also we need D4:

Stuff we thought was just plain wrong, but it isn’t.

Then there’s stuff we thought was wrong, then wasn’t wrong, then was wrong again.

We can take that much further. Several possibilities in astrophysics have bounced back and forth between possible and impossible at least half a dozen times. One is “wormholes”. Another is “cosmic contraction preceding the big bang”. A third, “more than three dimensions of space” has gone from impossible to possible to impossible to possible, and is now heading back towards impossible.

But more important is that many people in general and some scientists in particular refuse to admit that they were wrong. With the result that discredited ideas die only when people die. And that can take a long time.

Errors due to “allocation to wrong category” would exceed errors due to “insufficient papers examined” after about 300 papers were examined.

——————

On starting to go beyond 100 papers, my brain is starting to be less able to allocate papers to categories, perhaps that’s my limit – 100 papers in two days. One new interesting statement in a paper introduction:

“In the span of roughly five years, SDSS has more than doubled the number of known Milky Way satellites.” Wow! The paper goes on to show that one of the new satellite galaxies of the Milky Way (Laevens 1) isn’t a satellite galaxy at all, but a globular cluster.

You’ve heard of the “three-body-problem”, the problem that with three gravitational bodies not only is there no analytical solution, also the numerical solution blows up over a long enough timescale giving infinite results. I had known that the “three-body-problem” (and further the n-body problem) had been solved, but didn’t have any details.

One recent paper refers back to the solution method as: “The great insight of Wisdom and Holman (1991) was that, at the same level of approximation, one can add high frequency terms. By judicious choice of these additional frequencies, the perturbations among the planets can be transformed into trivially integrated delta functions. The result is an exceedingly efficient integrator that has proven an indispensable tool for modern studies in celestial mechanics”. ie. you can add in analytic instabilities that exactly cancel the numerical instabilities due to time-integration, giving reliable orbits an unlimited number of years into the future, or past.

mollwollfumble said:

You’ve heard of the “three-body-problem”, the problem that with three gravitational bodies not only is there no analytical solution, also the numerical solution blows up over a long enough timescale giving infinite results.

I have run more than a few computational analysis’ of this problem. It seems to me that the problem is trying to turn an analogue into a digital. Despite how far you divide the fluid motion of the analogue, you cannot in fact, specify exactly its location and vector, before both have changed.

Kingy said:

mollwollfumble said:

You’ve heard of the “three-body-problem”, the problem that with three gravitational bodies not only is there no analytical solution, also the numerical solution blows up over a long enough timescale giving infinite results.

I have run more than a few computational analysis’ of this problem. It seems to me that the problem is trying to turn an analogue into a digital. Despite how far you divide the fluid motion of the analogue, you cannot in fact, specify exactly its location and vector, before both have changed.

Me too “more than a few”. I agree.

One recent paper on spectra observed by Rosetta of comet 67P.
The outgassing from the comet is mostly hydrogen and oxygen molecules, created by dissociation of H2O on the comet surface by solar photons and electrons. Carbon is also observed in the spectrum, but only in the core – not in the outgassing.

One of the more data-driven papers is one about object IGR J17454−2919, which has steady state infrared and X-ray emission and transient X-ray bursts. At then end, the paper considers in turn the possibilities that it may be a neutron star, black hole, binary star, active galactic nucleus and blazar, ending rejecting all as unlikely. So WTF is it?

Here’s a fun paper. It describes how life could exist in subsurface environments living only off hydrogen generated by the breakdown of water when hit by cosmic rays from space. http://arxiv.org/ftp/arxiv/papers/1506/1506.01274.pdf

Kingy said:

mollwollfumble said:

You’ve heard of the “three-body-problem”, the problem that with three gravitational bodies not only is there no analytical solution, also the numerical solution blows up over a long enough timescale giving infinite results.

I have run more than a few computational analysis’ of this problem. It seems to me that the problem is trying to turn an analogue into a digital. Despite how far you divide the fluid motion of the analogue, you cannot in fact, specify exactly its location and vector, before both have changed.

I’d be very interested to hear more about your work on this, Kingy.

To see an example of how people doing numerics of the 3-body problem burden themselves with heavy pre-calculation analytics, see the ten pages of mathematical appendices of http://arxiv.org/pdf/1506.01382.pdf. Personally, I’m not convinced that that’s the best method.

Here’s some more trivia from a recent paper. It is peasantly warm (300 K) in Jupiter’s atmosphere at a pressure of a mere 5 bar. (Hot is 320 K at 8 bar). We could live at that pressure easily, given appropriate lung protection. Jupiter has four main cloudbase levels of 0.7 bar, 2.2 bar, 5.4 bar and 5.7 bar. So we’d be within the clouds in regions where the air is being convected upwards.

mollwollfumble said:

Results from 100 arXiv (first hundred from June 2015). A hundred is nowhere near enough for a reliable analysis of this sort, because the timing of articles placed in arXiv is far from random. Perhaps best to add a +-5%.

Sturgeon’s law is “90% of everything is crap”. But is it really 90%?

Next 100 arXiv (number 101 to 200 from june 2015). This time 45% of papers remained uncategorised because of brain malfunction.
Results. Combined%, (First 100, Second 100).

A) Crap science (15.5%)
6% (5% 7%) 1) Papers on ramifications of scientific hypotheses that have already been disproved (including those rejected by Occam’s razor).
4% (1% 7%) 2) Observations that are not statistically significant.
2.5% (3% 2%) 3) Hobby horse science – the same paper published over and over again with only trivial modifications.
1.5% (2% 1%) 4) Minimal effort, would have required less than three man-days to research.
1.5% (2% 1%) 5) Just plain wrong.

0% (0% 0%) E1) I haven’t got a clue what they’re talking about.
(0% 45%) E2) Unable to categorise without more effort

mollwollfumble said:

Results from 100 arXiv (first hundred from June 2015). A hundred is nowhere near enough for a reliable analysis of this sort, because the timing of articles placed in arXiv is far from random. Perhaps best to add a +-5%.

Sturgeon’s law is “90% of everything is crap”. But is it really 90%?

Next 100 arXiv (number 101 to 200 from june 2015). This time 45% of papers remained uncategorised because of brain malfunction.
Results. Combined%, (First 100, Second 100).

A) Crap science (15.5%)
6% (5% 7%) 1) Papers on ramifications of scientific hypotheses that have already been disproved (including those rejected by Occam’s razor).
4% (1% 7%) 2) Observations that are not statistically significant.
2.5% (3% 2%) 3) Hobby horse science – the same paper published over and over again with only trivial modifications.
1.5% (2% 1%) 4) Minimal effort, would have required less than three man-days to research.
1.5% (2% 1%) 5) Just plain wrong.

0% (0% 0%) E1) I haven’t got a clue what they’re talking about.
(0% 45%) E2) Unable to categorise without more effort

mollwollfumble said:

Here’s some more trivia from a recent paper. It is peasantly warm (300 K) in Jupiter’s atmosphere at a pressure of a mere 5 bar. (Hot is 320 K at 8 bar). We could live at that pressure easily, given appropriate lung protection. Jupiter has four main cloudbase levels of 0.7 bar, 2.2 bar, 5.4 bar and 5.7 bar. So we’d be within the clouds in regions where the air is being convected upwards.

don’t you have heaps of radiation coming off Jupiter?

plus venus would be better for the first jump off

its closer to the sun – solar panels work better but it could be hazy as its sort of in the cloud

its roughly at 25 deg c at 1 bar

it takes less time to get there – important if a rescue mission needs to be mounted

wookiemeister said:

mollwollfumble said:

Here’s some more trivia from a recent paper. It is peasantly warm (300 K) in Jupiter’s atmosphere at a pressure of a mere 5 bar. (Hot is 320 K at 8 bar). We could live at that pressure easily, given appropriate lung protection. Jupiter has four main cloudbase levels of 0.7 bar, 2.2 bar, 5.4 bar and 5.7 bar. So we’d be within the clouds in regions where the air is being convected upwards.

don’t you have heaps of radiation coming off Jupiter?

Do you? You have some fierce lightning, but that’s about all. The “heaps of radiation” is out in the ionosphere, the Van Allen belt. I’m not exactly sure how far out the worst of the radiation is, but it’s probably unwise to hang around on the surface of Io.