Hacker News Reader

Comment by riahi | Hacker News Reader

You have to build the background knowledge to even be able to speak the language of medicine. This is not difficult "just because", it's difficult because it's a huge amount of baseline knowledge that must be assimilated, before you can even begin to think critically.

This is like telling someone "we need to figure out how to use technology to prevent people from spending years of rote memorization before they can speak <X_foreign_language>!"

View on HN

You can undoubtedly put priority of grammar over vocabulary, no? You are not thinking in terms of systems that allow us to find the "words" we need after being guided by our mastery of the grammar and ability to reason about what must be said given a situation. I think what you said is valid only if we think about speaking a language unassisted by these types of systems, which is exactly what I feel must be left behind.

Edit:

Also, this is what happens in practice anyway! Doctors don't remember a lot and constantly have to look things up ( as is understandable and expected ). You must be a master of the "grammar of medicine" though. Otherwise you won't be able to construct well formed sentences... aka medical thoughts.

You need some memorization to give some concrete instances of the grammar, of course; however, going overboard on the "vocabulary" is not constructive and shifts the focus onto knowing a lot of words, not knowing how to construct great sentences. It's like a writer that knows a lot of uncommon words, but ends up having nothing meaningful to say... because the power of writing is not in the vocabulary alone, but in the craft of composition.

> Doctors don't remember a lot and constantly have to look things up ( as is understandable and expected ).

They don't bootstrap this search every time. Enough of the basics have been instilled that they will know what they should look up.

The problem with that analogy is Med school treats medicine like speaking English. But you can simplify the material and treat it like teaching a constructed language.

For example if you named the bones in the fingers as left, thumb, two then it’s little effort to recall. Instead people need to recal the stuff like distal phalanges. Math is arguably worse with redundant notations for the same thing. So, I see the benifit of avoiding creating multiple systems.

I don't think that's a good example. There are only 206 bones in the adult human body. It doesn't take very long to memorize those. The real memorization time is spent on other stuff for which we don't even have common English names. You could invent new, simpler names for every anatomical part, every disease, every symptom, but that would be a huge undertaking and would only make memorization marginally easier.

My point is if some of med school is obvious cruft that can be removed and or altered to be more useful in practice. Why not?

PS: I would include tendons, joints, and muscles etc not just 206 bones in that example. Also, that was rather off the cuff as left/right could confuse things so maybe you want port vs starboard or something.

You’re really displaying your total ignorance of medicine now - our anatomical naming conventions are highly specific. Anterior, superior, distal, medial, ... hell, even phalanx means finger. Your lack of knowledge is causing you to make very general assumptions about the teaching and learning of medicine that are incorrect

Phalanx does not mean finger to most people until they start learning medicine. It’s cruft.

Phalanx also means finger to most students of history. It is, after all, Greek for Finger (secondary to Aristotle dubbing the bones so).

What's your point? It seems to be that medical terminology (or the expert body of knowledge that distinguishes someone with anatomical or medical knowledge from the layperson) is an artificial construct which imposes undue restrictions on the easy transfer of this knowledge - guild behaviour.

Every realm of knowledge has its own language which must be learned in order to be proficient. In briefly reading your other replies in this thread, it seems like you feel that computing and programming don't have these restrictions - that is absolutely not true, just ask anyone who has been trying to learn programming themselves but is stuck on the difference between a method and a function (only to find out they are the same thing).

It is not possible to be an expert without gathering the knowledge, the language, and the real understanding necessary to partake in the field. Anything else is simple arrogance, to assume that any specialised field that people spend decades learning, is actually just arcane crud and cruft that disguises the simplicity hidden within

I think you’re getting the wrong impression. I find medical papers tend to be easier to read as a non specialist than recent matimatics papers.

Math however is trying to be as clear as possible and more elegant notation generally though not always wins over time. Mathematics don’t still use Roman Numerals even though they worked.

PS: Languages change over time, programmers dropped the term subroutine. This can look messy, but it’s benifical over time.

I may be getting the wrong impression - but what is the point you’re trying to make - and did you just choose a shitty example? The left, thumb, Two, carries so many problems - do you mean palmar or volar surface? Medial or lateral? I can accurately discribe a lesion at the base of the left thenar eminence and everyone who has studied anatomy anywhere in the world will understand the exact part of the body I am talking about. In another more specific example, I might describe petechial haemorrhages proximal to the iliocaecal valve. These words aren’t in common usage but again, every medically trained human in the world would know exactly what I am describing.

I understand your last comment about languages changing over time (technical languages) - medicine is no different. See for instance the Dukes grading of bowel cancer, now largely superceeded by TNM Grading which carries specific histological and pathological meanings (as dukes does) but is applicable to all cancers.

Yeah the idea is skip names as the foundation altogether. Instead characterize the spaces by their properties and symmetries, and when there's an exact number of things (like bones) use a GUI (body diagram, decision tree) to air recall so people can just learn the names on the job.

>so people can just learn the names on the job

Ah, yes, the best way to learn all the intricacies of a human: osmosis.

Good luck attaining knowledge that takes 8+ years of focused study in medical training today through osmosis over 30 years of "on the job" training, being paid $0 because you are still "learning" and can't contribute to anyone's actual medical care during that time unlike a resident physician.

They did what you suggested.

“Distal phalanges” are the far away phalanges, which are the bones that are in an array like a phalanx in the extremities.

The problem is that language changed after that, and the expectation for those words to make intuitive sense changed in the broader culture doctors were pulled from.

They’re then faced with having to rename everything, breaking with the established record and creating confusion or having to explicitly teach the names.

I’m also curious what you think is redundant in math notation.

https://en.m.wikipedia.org/wiki/Notation_for_differentiation

Taking classes that use different notation for the same thing is annoying.

Your example is not analogous to medicine - there is consistent and highly specific naming conventions for a reason

It’s not generally ambiguous in math, and a similar situation exists when a patient describes his symptoms using different words.

That said, I started with: ‘Math is arguably worse with redundant notations for the same thing.’

He asked “I’m also curious what you think is redundant in math notation.”

Although I would welcome the use of common names instead of Latin names, the names in anatomy and pathology are annoying but not a hindrance. The difficult thing is to remember causes and symptoms of diseases, innervation of muscles, drug pathways, etc. That would be hard to simplify.

Instead of Latin/Greek root words, you'd have to learn a new system of common words. And not just new trainees, but also practicing physicians, because everyone has to speak the same language.

All fields have jargon, but math culture is at least "up to isomorphism".

Reply on news.ycombinator.com