Take the example of an X-ray. A skilled physician will see a lot more in an X-ray than a layman. The picture absorbed into the layman's head is in some sense a totally different picture than the one absorbed into the physician's head, even though the source object is the same.

The "source object" cannot ever just by itself create an image in anyone's head; the image creation happens because the viewer's brain is expending effort in absorbing it, and in the process applying all sorts of unconscious filters that are a result of their brain's unique design.

In the same vein as above, theories do not come from just facts, they come to a large extent from other theories, and facts can thereafter be observed to possibly falsify the theory. However one reason you may not succeed at falsifying it, is when the facts you observe are in part affected by the theory!

More concretely, what enters into the set of facts that ultimately gets observed by you is affected by the sum of your theories – even your world-conception – analogous to how the skilled physician comes away from the X-ray with a different set of facts than the layman does. If she'd been trained by cargo-culting witch doctors, she'd again come away with a spectacularly different set of facts.

To use a another example, if you are 100% certain that ghosts don't exist, you will never experience seeing a ghost even if an actual ghost enters your field of view, because when this happens, you presuppose it is not a ghost but something else.

Closer to recent affairs, if you've bought into the arguments of Big Tobacco, all data about smoking seem to argue that it's harmless. Or if you're a 90s student of nutrition with no reason to doubt your textbooks, the world seems to be overflowing and exploding with evidence that dietary fat does indeed harm people's health. Yet now the community is questioning if we ever had good data for that.

This is one reason why it's important to have a well-calibrated model of the world before you even create a new hypothesis to look into. (Calibration) There are even mathematical lemma in statistics about this principle.

Beyond that, we can aim to have more of a self-correcting system of beliefs, which starts with good epistemology: understand Meta-science and root out patterns of Anti-epistemology.

Self-correction also seems to require a certain self-distance… an ability to cope and work with the fact you may be wrong right now, which may be composed of at least two things: an ability to model yourself as running on malicious hardware (Cognitive science), and the fundamental acceptance that things can't be argued into being true but that we only ever argue ourselves towards the truth or away from it, which remains as fixed as a star in the sky, and equally disinterested in distinguishing itself from other hypotheses as a star is from other stars (Rationalization is irrational). The understanding that when you believe something, it won't feel like you "believe" it, it'll feel like it just is the case. That understanding that gives you the ability to reason about this feeling as if it was indeed only someone's belief (where the "someone" happens to be you) and not the truth even though internally to you it's Obviously Just Simply True. I'm told not everyone has that ability, but all my science & science fiction heroes did and I can't imagine what it'd be like to not grow up to revere it and seek to emulate it, but it sort of explains why not everyone has it, since it does sound difficult when written out like that.

Some philosophers debate about whether facts can be gotten from your senses… at all! Say what? There is an array of situations where it's probably valid to suppose yes. Counting ticks on a Geiger counter and reading measurements on a thermometer are unproblematic examples. But you can argue that a large part of our world is actually not facts but constructions, thus there are fewer universalities than we'd think – we need a shared culture to make more universalities.

It seems meaningless to say a field of science is "fact-based" – if we make claims based on observations, do these claims themselves count as new facts? No. Then what is meant? An applause light, probably. If you contrast astronomy with astrology, proponents of both will try to say their field is fact-based, the only thing at question is whether they succeed in being so. I.e. whether they sensibly link the facts with the claims they support. It is not a question whether they try to be fact-based at all, because of course everyone tries.

How do we link facts with claims? There's usually a chain of ever more indirect facts. The golden standard is when a small set of facts can prove the truth of a convenient abstraction which in turn supports many new facts; the chain from the original unproblematic facts to any one of these will be short. Many of these latter facts are properly considered theory-laden observations. That is, we have facts, then a theory inspired by the facts, and the theory then allows us to interpret in a new light what we see, giving us new facts. But the latter facts are facts only if the theory is true.

Astronomers are probably familiar with going back to reinterpret lots of old data after an update to astronomy. The theory has changed, so it's possible some "facts" should be forgotten and some new ones brought up. This seems relatively easy in astronomy, because your raw data is always in a clearly marked bucket and you don't reuse it in everyday life and mutate it…

• What does this mean for my world-conception?

• Objectivity
• My Copia
• Book: What Is This Thing Called Science?

• Computing

Searching the web for a timeline of geologic periods, most of the infographics could be condensed to a single 10-point bullet list. Here are some examples that are at least a bit more interesting:

• 1 – this one has maps of how the continents looked at each period – a great addition to any infographic. However, the phylogenetic info is superficial, and being just pictures, hard to look up.
• 2 skmonug.blogspot.com/2020/04/figure-79-sketch-of-evolution-of-plant.html
  • this embeds detailed phylogenetic information, but just for one group of life ("plant forms")
• 3
  • Close but no cigar - the squiggly lines seem to be of artistic origin rather than mapping to known genera, the bacteria part should be as big as the eukaryote part, not all extinct genera are shown etc. But the concept is exceptionally well done.
• 4
  • not about geology at all, but very creative!
• 5 www.dailymail.co.uk/sciencetech/article-2982564/Tree-life-rewritten-Evolutionary-timeline-looks-like-giant-lollipop-reveals-new-species-appear-2-million-years.html
  • yes! this is what infographics are about, enabling new ways to think
• 6 timetree.org/ – a data source
  • good poster at timetree.org/book but full res needs $$$
• 7
  • great! it explains a bit e.g. organs, bilateral symmetry, arthropod "joint-foot", primates "first", first opening in cell cluster becomes mouth…
  • you can tell that this one was made by an university student or teacher for learning/memorizing

Opinion piece

Almost every world map I've seen was basically just showing where each country is, plus maybe one other dimension of info (like GDP or temperature or topography).

Even "physical" maps don't succeed in showing much beyond topography and general climate, and there is often still an overwhelming focus on borders. And those that remove borders, don't take advantage of it, resulting in a info-poor map. Large regions appear to just be blank. Famous rivers like the Elbe are not shown. Maps that do succeed at showing things like the Elbe then proceed to not show any cities, so I can't see where e.g. Hanover is in relation to the Elbe. Come on!

Then you look at artistic maps, but then they seem to only care about making an artwork and forget to embed much info. Why can't a map be both artistic and a good place to look things up?

I'm also bothered by inappropriate coloring, e.g. when a "physical map" seems to say something about climate with its colors but actually is just colored by altitude, leading to the appearance that there is no greenery in high-altitude areas. I'd rather areas be colored more true-to-life.

The question of how to show altitude can be interesting, because I for one have rarely found absolute altitude enlightening. It's difference in altitude that would be more useful to know, and how mountainous an area is (how "spiky" the altitude), in a way that leads to actual difficulty to walk/bike through. In practical life, we don't recognize mountain by "height from sea-level", but by "height from surroundings".

One probably good way to deal with that is to use altitude lines, or by just artistically drawing mountains (as in fantasy maps) in areas where you would in practice notice them.

Don't show these on a map for God's sake

• Road and rail (these belong to the era of phonebooks, road maps and manual navigation)
• Borders
• Country names

Show more of these, please

• Rivers
• River names
• Small towns
• Named geo-features like peninsulas, valleys and forests – just try to look for Schwarzwald on a modern map, good luck!
• UNESCO sites
• Cultural stuff (not sure what)
• Frequently-active volcanic cones (fun for kids)
• persistent sea-surface currents
• marginal seas (e.g. Ligurian Sea)
• Some districts of big cities (not just a big "London" blob)

As a check, at a minimum, a map worth a cent should succeed at naming all of the following:

• various basins of the Sahara
• some towns inside the Sahara
• the Horn of Africa
• the Siberian Traps
• Honshu Island
• Torneälv
• Vindelälv
• The Elbe
• Birka
• Edinburgh of the Seven Seas
• The White Sea
• The Ligurian Sea
• Schwarzwald

A map should NOT bother to name

• any continent (everyone's taught the continents at age 7)
• any country (but maybe show island ownership)
• The Atlantic or Pacific oceans
• The Sahara

not sure if it should be a Dymaxion map, but it would be cool, esp. if it came in pieces that could be assembled into a rough-hewn sphere, or posted flat on the wall as you prefer

• www.bigocheatsheet.com/
• old.reddit.com/r/dataisbeautiful/comments/1apqsi9/oc_timeline_of_geopolitical_changes_in_europe/

• 2022-01-25