Well at least one can talk together about stuff and then see if the other suffered and dislikes certain parts or maybe finds it great and exactly why. Then you know quite fast how many experience he/she has in the field.
In the end there is just paint/words. The question how to find and apply the right/wanted paint/words involves many aspects. By training one aspect it doesn't improve the other, but it may or may not improve the overall outcome.
Imho he did not be became better at "painting" but he made better designs for paintings. He trained one important aspect of the whole piece of art, the design. The aspect of applying paint was not trained, so that probably has not improved.
The question is does the person performing the job him/herself finds it usefull what he/she does. Is doesn't matter if it is useful in your eyes. Because it has to do with appreciation, selfconfidence identification, finding satisfaction. If your are just being giving the feeling to be stupid useless replacable person by letting you jump trough useless hoops to please undefinable goals for more powerfull and this happens in a statistical signifant amount
then it tells something about current state of society and gives you some questions: why ist that ? , how do we handle the other humans around us?, what is causing that mindset and in what future, how would you/we like to live together?
Nice but the most interesting is left out imho. It stops at the human retinas. But the way color then triggers perception would, I assume, also have a very high influence.
For example we relate warm and cooler colors to turning forms relate the distribution of brightness and saturation to form, distance etc. Then the precieved forms and colors are related to feelings. So a better suitable colormap for medical images would require more to prevent misinterpretations.
'Pretty' is maybe also subjective but probably not arbitrary...
Well the fact that the companies own the stuff that was researched with tax payer money, which then tax payer money must buy again also doesn't help to scale. As there is no real market but just an oligopol of producers. If the formulas would be public domain, the producers can compete who is fastest and there would be much more and they still earn good money...