Why do doctors memorize medicines and their effects? Because it's faster to retrieve from memory than from documentation, even automatically presented documentation. It removes a context switch.

The "brain cycles" are only run while doing the memorization, not while actually doing the work.

> I don't recall ever struggling to remember syntax of languages I frequently code in.

Because you've memorized it.

You definitely have no idea of how doctors work if you think they don’t heavily rely on reference books for most of the stuff they do

I don’t think you’re being generous with your reading of their point.

You can't easily consider a treatment or remedy or library function you don't remember. Better to use all the tools at hand (memory being a useful tool - a L2 cache as it were) than handicap yourself by only using a few.

I'd rather have my doctor consulting recent knowledge base than use their 30 year old memories from school. Medicine is not set in stone and recommendations change.

As I've heard it, use your memory, then back it up with documentation if/when you need the certainty.

You can't easily consider a treatment or remedy or library function you don't remember.

The problem with using cache is that you can feel certain it's correct but you don't really know until you check with the source. And since you check with the source anyway, the cache becomes a drag to correctness

Do they have every single medicine memorized, or just the ones they most commonly use? Sure, I tend to remember the parts of libraries and apis I use often, but I'd still argue my editor having a reminder just a keystroke away is very beneficial.

Reality: They rely on medical representatives to know what the actual medicines are. They remember the ones with the most kickbacks.

More than just the most commonly used...

https://www.youtube.com/watch?v=jR7d56UeeeU <= Song-based memorization aide for 200 different medications

There are over 20,000 approved medications in the US. I don't really feel like this analogy is holding up very well. But, have I happened to remember about 200 or so API and library calls over my career? Probably.

Yeah the idea that I have to remember everything in our codebase of millions of LoC made up of TS, JS, Ruby, Kotlin, and Java is ridiculous. IDEs allow me to drop into these codebases and be decently productive from the get go.

When you start to work with too many languages frequently, using syntax from one while writing another is common and IDEs really help at this point. I too prefer not to remember APIs and methods, but with time, our brain starts maintaining a cache of these. Completions make this faster.

That and error detection are only features I like in IDEs. For everything else, dedicated tools are often better (but not necessarily easier) than integrated ones.

Do you memorize all those IO/XML/JSON/Hadoop (HDFS/MR/Core/...)/AWS/Azure/... APIs by heart?

The parent explicitly said they do not want to memorize APIs.

I understand, I just wonder if he/she really means to it. Who knows, some people have very good memory. When I work with a new (for me) topic, I am trying to get basic understanding of topic, still it far from remembering exact method signatures.