Hacker News Reader

Comment by laurentlb | Hacker News Reader

There are many ways to use LLMs.

The issue, IMO, is that some people throw in a one-shot, short prompt, and get a generic, boring output. "Garbage in, generic out."

Here's how I actually use LLMs:

- To dump my thoughts and get help organizing them.

- To get feedback on phrasing and transitions (I'm not a native speaker).

- To improve tone, style (while trying to keep it personal!), or just to simplify messy sentences.

- To identify issues, missing information, etc. in my text.

It’s usually an iterative process, and the combined prompt length ends up longer than the final result. And I incorporate the feedback manually.

So sure, if someone types "write a blog post about X" and hits go, the prompt is more interesting than the output. But when there are five rounds of edits and context, would you really rather read all the prompts and drafts instead of the final version?

(if you do: https://chatgpt.com/share/6817dd19-4604-800b-95ee-f2dd05add4...)

View on HN

FWIW: Your original comment, in the first message you sent ChatGPT, was way better than the one you posted. Simple, authentic, to the point

I couldn't agree more, this 'polished' style the finished comment comes in is super boring to read. It's hard to put the finger on it, but overall flow is just too... Samesame? I guess it's perfectly _expected_ to be predictable to read ;)

And then one of the iterations was asking additional ways llms could be used and then adding some of those as content which seems odd but plausibly helpful brainstorming. Just the phrasing of the original question makes it sound like things the user isn't actually doing but wants in their comment if that makes sense.

Thanks for the example chat it was a valuable learning for me!

>- To improve tone, style (while trying to keep it personal!), or just to simplify messy sentences.

Ironically enough, as I was reading your post this is what convinced me it was written by ChatGPT.

FWIW, the initial draft you gave to chatgpt is better than what you posted.

That's a lot of back-and-forth to produce something that's functionally equivalent to your original text, only slightly worse.

> would you really rather read all the prompts and drafts instead of the final version?

I think you missed the point of the article. They did not mean it literally: it's a way to say that they are interested in what you have to say.

And that is the point that is extremely difficult to make students understand. When a teacher asks a student to write about a historical event, it's not just some kind of ceremony on the way to a degree. The end goal is to make the student improve in a number of skills: gathering information, making sense of it, absorbing it, being critical about what they read, eventually building an opinion about it.

When you say "I use an LLM to dump my thoughts and get help organising them", what you say is that you are not interested in improving your ability to actually absorb information. To me, it says that you are not interested in becoming interesting. I would think that it is a maturity issue: some day you will understand.

And that's what the article says: I am interested in hearing what you have to say about a topic that you care about. I am not interested into anything you can do to pretend that you care or know about it. If you can't organise your thoughts yourself, I don't believe that you have reached a point where you are interesting. Not that you will never get there; it just takes practice. But if you don't practice (and use LLMs instead), my concern is that you will never become interesting. This time is wasted, I don't want to read what your LLM generated from that stuff you didn't care to absorb in the first place.

* I think you missed the point of the article. They did not mean it literally: it's a way to say that they are interested in what you have to say.*

Maybe an LLM could have helped the writer say what they meant to.

Maybe if people were more used to reading and less to copy-pasting into LLMs, they would understand the article.

If your readers don’t understand what you’re writing, it’s your fault as the author. There’s so much good stuff to read, why waste time on someone who can’t organize or articulate their thoughts?

Exactly it is a tool which needs skill to use. I would add extra use of mine:

- To "Translate to language XYZ", and that is not sometimes strightforward and needs iterating like "Translate to language <LANGUAGE> used by <PERSON ROLE> living in <CITY>" and so on.

And the author is right, I use it as 2nd-language user, thus LLM produces better text than myself. However I am not going to share the prompt as it is useless (foreign language) and too messy (bits of draft text) to the reader. I would compare it to passing a book draft thru editor and translator.

For what it's worth, I think that sending a message translated to a foreign language you don't master is the worst thing you can do.

You speak English? Write and send your message in English. The receiver can copy-paste it in a translator. This way, they will know that they are not reading the original. So if your translated message sounds inaccurate, offensive or anything like that, they can go back to your original message.

Reply on news.ycombinator.com