Hacker News Reader

Comment by Joker_vD | Hacker News Reader

Heh. Mine, before the diploma defense, literally told me "Assume that the panel knows everything about the math except for the stuff your did in your thesis, so tell them about that. Don't bother explaining what 'error correction codes' are, tell them about your specific construction and practical boundaries you've determined". Which, I guess, is precisely "the reader knows everything I know except this one thing" antipattern but then again, blogging is different from e.g. conference presentation, isn't it?

View on HN

Yeah, deciding on your assumptions about the audience's prior knowledge is one of the toughest parts of presenting information well, I think. If you give too much info, you come across as boring or patronizing; if you don't give enough info, most (if not all) of your audience will rapidly lose interest. But it's all gotta be based on your exact audience in the moment, and that's really hard to take into account a lot of the time.

> Yeah, deciding on your assumptions about the audience's prior knowledge is one of the toughest parts of presenting information well, I think.

Right, all good (human) communication boils down to both the author and the reader having a theory-of-mind for one-another. The writer must anticipate what a reader will be thinking, and a reader must imagine [0] what the writer wanted to convey.

That's part of what makes the modern epidemic of LLM slop so frustrating: The correctly-formed words make us waste effort trying to reconstruct a mind that was never really there. Like some animals encountering an extremely realistic plastic fake that we cannot eat/mate-with.

Ooh yes, I like that way of framing the problem. I'll have to remember that!

When writing text, you have the advantage of being able to link to another text. In principle academic citations form a DAG that allows you to understand all prerequisite knowledge for any given paper. Your paper is expected to cite everything it depends on that isn't common knowledge but doesn't need to re-explain it.

Well, not necessarily a DAG — a colleague in my field managed to get two submissions accepted at the same conference that cited each other. I'm sure somebody's publications crawler was upset about this at some point...

I remember a smartass kid in year 12 that cited his own essay in the same essay and got points for it being cited properly, but points off for not being a primary source.

I have found the opposite to be true in conferences, but it might be different for maths than biology?

The worst presentations are always the one that assume that you know every experimental technique and every bit of knowledge. I always took the stance that the audience had at most a master's degree in the field and took care to explain everything, even briefly.

Papers are written to peers. It shouldn't be a "murder mystery" for them (that doesn't mean it shouldn't tell a story). You have to write to your audience. If a paper was written to a general audience, instead of peers, every paper would have to be a novel.

I'll admit that sometimes I'm frustrated with HN comments on arxiv articles. Despite being a nerdy forum most papers aren't written for this audience. Hell, I have a PhD and most papers aren't written for me. Nor should they. I'm perfectly okay if I don't understand a paper. Arguably the hardest thing about a PhD is learning to be comfortable with being dumb and willing to work on becoming less dumb (you're never not dumb, only less dumb)

That's true, but I think scientific writing often veers too hard towards impenetrability. At least in machine learning, many of the hallmark papers are remarkably clear and simple in a way that most papers I read day-to-day aren't. Even something as simple as the problem setting can be written insightfully in a strong piece of work.

  > At least in machine learning

You should read math papers lol.

Honestly, I feel the opposite. But maybe that's because my doctorate is in ML. But we might also have a different definition of what the hallmark papers are. Many of the well known papers are ones that model scaled an existing method (ViT and ddpm are good examples). Important papers, but in which sense? I do think we have space for all types of papers and written to different audiences, but I don't think we should punish papers just because they're written to their niche domain. In the past decade there's been a huge increase in citation farming and I don't think that helps. Math is probably too much on one extreme but I think ML is on the other. You have to write your papers so a first year PhD can understand, because they're the ones reviewing your work. I don't think that's a desirable outcome

I feel that this partly due to the fact the these famous/groundbreaking papers will usually be explained to you in many other places before you even read it. So you probably get good understanding of it even before reading the actual paper.

For example, the papers where Einstein presented special relativity suck really but most of people know about it with goor grasp before reading the actual papers (People of that era).

Reply on news.ycombinator.com