Hacker News Reader

Comment by aappleby | Hacker News Reader

I am utterly baffled that no response in this thread so far has taken issue with the statement "As you probably know intuitively, it won't work."

For me, this _does_ work and I have proven it many many times over the years by adding entire categories of technical knowledge to my repertoire. And not superficially, either - I get paid very well to do things professionally that I taught myself by reading Wikipedia.

If my experience were commonplace, the "it won't work" statement would be highly contentious in the comments here. Since it isn't, I guess I can deduce that I must be an outlier.

View on HN

Maybe you're really good at working with uncertainty?

This article focuses on the number of new things overwhelming our working memory, but I'd argue another big problem is the difficulty of leveraging abstractions even if you don't understand them.

In this exercise, both myself and the article get sidetracked in understanding the basis of all the abstractions as well, rather than just letting the abstractions be, with all of their uncertainty. But if I were better at the latter, I could see this approach working.

I did the exercise mentioned and was happily bouncing around between 3 and 6 levels deep in Wikipedia before I thought I should come back to the article to see what else it said, at which point I became very confused.

Perhaps "comfortable with knowing that I don't know things, yet proceeding anyway" is a better description.

Maybe you already know quite a lot about physics? Or maybe you have an unusually large working memory and didn't go far enough down the wikipedia hole until you got lost?

The working memory limit explained in the article must apply to you unless you have some sort of a very special form of photographic memory

That's the thing though, I never get "lost".

I've never studied quantum mechanics aside from snippets I picked up through pop science articles or such, so it's not like I started with a big body of knowledge already in place.

It's hard for me to imagine what feeling "lost" in Wikipedia would be like. Maybe reading some article about a complex historical battle would do it - lists of people with important names would probably become a jumble without really clear identifiers of who did what and why.

Ok how about this more difficult experiment to prove the hypothesis:

I'm going to send you a wikipedia page written in chinese. You're going to read it once. For every chinese word you're allowed to look up the meaning on google translate once as you read. You're not allowed to do any other form of learning besides reading.

Do you think in this case you'd also not get "lost"?

I think I would get lost, but only due to the vocabulary and translation load and the "once" restrictions - not due to the content of the article itself.

In practice I end up reading the same wiki articles dozens of times as I go up and down the link trees, with each pass connecting various ambiguous ideas together.

If I could annotate the pages with arbitrary amounts of translation notes as I read them, I could probably get through the article though it would take a very long time.

Ok but the point stands that there is a limit to your working memory. Your limit sounds a lot higher than most peoples but it is still limited otherwise you’d never get lost even if you could only read things once.

I don't _feel_ like my working memory is particularly large, but that's also a hopelessly subjective assessment.

Maybe it's more like my working memory can hold more things by aggressively discarding details until I need them - same capacity, but with better lossy compression. Not sure if I could measure that either, but it feels closer.

Selection bias. People read “learning is memory” and choose not to waste their time reading the piece. If the headline is silly, usually the rest is too.

Reasonable, but it's literally the first point the article makes. Of the fraction of commenters that did actually read the article, I still would've expected at least a few to object.

Reply on news.ycombinator.com