4 ms·

This post is a bit messy. The author starts talking about ads and algorithms manipulating us before quickly saying that subscriptions are similarly harmful, exc

This post is a bit messy. The author starts talking about ads and algorithms manipulating us before quickly saying that subscriptions are similarly harmful, except that link is not really explained. It's randomly shoehorned in. The author likens John Deere and Netflix but quickly brushes aside the fact that digital media and physical tractors are very different things to own. Finally, the author suggests that paid subscriptions are bad because they can restrict our access to material and steal our data. I concede the first point but not the later. Reading my news subscriptions is significantly better than reading "Google News" and other ad-driven media suggestions. Overall I am disappointed in the quality of the argument. This person does actually have a PhD which makes me want to talk to this person in a long form conversation where they won't feel constrained by this specific medium (pun intended).

Handprint4469 3y ago
Out of curiosity, did you use ChatGPT to help you write this comment? It uses the following format: > The author starts talking..... The author likens..... Finally, the author suggests...... Overall, I am disappointed in.... Which, while a perfectly valid way of structuring a comment, it's more usually found in ChatGPT-style answers than in random comments on the Internet.
- LargeTomato 3y ago
  I promise you I did not! But I have recently started changing my "prose" to be more clear, layering information thoughtfully, and spending more words on filler (e.g. The Author Suggests). I definitely learned part of it from reading chat gpt. I like it better than some of the more math-y or academic writing which I will sometimes use but only if I need to be hyper specific.
  - jeroenhd 3y ago
    I fear for the future where every comment that has been composed with some thought will be branded as ChatGPT generated. I know most of those suspicions will be correct if they aren't already, but I'm still not very happy about it.
    - bityard 3y ago
      It's not the amount of thought that brands it as something that could be mistaken for a chatgpt emission. Rather, chatgpt writing has a very formulaic style that is hard not to instantly recognize once you have seen it a few times. It's not super surprising to me that people who use it frequently begin to unconsciously imitate the chatgpt style in their own writing after a time.
      - LargeTomato 3y ago
        It makes my job at writer much easier. If I don't want to think about prose I can copy chat gpt and get 80% of the way there with very little effort. Most stuff I write is for my coworkers or Hacker News so I don't mind. When I write for fun or for my family I like to be more thoughtful.
  - bityard 3y ago
    Congratulations, you have passed the inverse Turing test
  - WirelessGigabit 3y ago
    I don't like it. Now I need to parse much more before I get to the interesting part. It seems that ChatGPT learned from Recipes webpages. Include a story on how the author's parents had sex on a beach in Maui which sparked the author's desire for Mango Passion Fruit Cakes. In fact, just like I have a limited amount of keystrokes in my lifetime (which incidentally I'm spending on this post) I have a limited amount of time to go through text. Soon I'll start skipping the longer ones.
    - LargeTomato 3y ago
      I used to agree with you, and to an extent I still do, but being able to carry the reader through the text by providing filler makes my writing appear more clear even if it's actually has lower information density. I do this with speaking, too. I used to be a bit of an abrasive engineer and just "tell it like it is" but after hitting a communication wall I started to coddle the listener and people really like it. I personally think it's unfortunate but it is what it is.
- sragh 3y ago
  That style is also totally in line with correspondence into magazines like The Economist and Time. It’s not exclusive to ChatGPT
  - LargeTomato 3y ago
    I love the Economist and I have gotten a lot of positive feedback on my writing whenever I try to adopt their tone.
  - tkgally 3y ago
    It’s not very common in HN posts, though. I also suspected assistance from ChatGPT. It’s interesting to read that the author is consciously learning from it.

Handprint4469 3y ago

Out of curiosity, did you use ChatGPT to help you write this comment? It uses the following format: > The author starts talking..... The author likens..... Finally, the author suggests...... Overall, I am disappointed in.... Which, while a perfectly valid way of structuring a comment, it's more usually found in ChatGPT-style answers than in random comments on the Internet.

LargeTomato 3y ago
I promise you I did not! But I have recently started changing my "prose" to be more clear, layering information thoughtfully, and spending more words on filler (e.g. The Author Suggests). I definitely learned part of it from reading chat gpt. I like it better than some of the more math-y or academic writing which I will sometimes use but only if I need to be hyper specific.
- jeroenhd 3y ago
  I fear for the future where every comment that has been composed with some thought will be branded as ChatGPT generated. I know most of those suspicions will be correct if they aren't already, but I'm still not very happy about it.
  - bityard 3y ago
    It's not the amount of thought that brands it as something that could be mistaken for a chatgpt emission. Rather, chatgpt writing has a very formulaic style that is hard not to instantly recognize once you have seen it a few times. It's not super surprising to me that people who use it frequently begin to unconsciously imitate the chatgpt style in their own writing after a time.
    - LargeTomato 3y ago
      It makes my job at writer much easier. If I don't want to think about prose I can copy chat gpt and get 80% of the way there with very little effort. Most stuff I write is for my coworkers or Hacker News so I don't mind. When I write for fun or for my family I like to be more thoughtful.
- bityard 3y ago
  Congratulations, you have passed the inverse Turing test
- WirelessGigabit 3y ago
  I don't like it. Now I need to parse much more before I get to the interesting part. It seems that ChatGPT learned from Recipes webpages. Include a story on how the author's parents had sex on a beach in Maui which sparked the author's desire for Mango Passion Fruit Cakes. In fact, just like I have a limited amount of keystrokes in my lifetime (which incidentally I'm spending on this post) I have a limited amount of time to go through text. Soon I'll start skipping the longer ones.
  - LargeTomato 3y ago
    I used to agree with you, and to an extent I still do, but being able to carry the reader through the text by providing filler makes my writing appear more clear even if it's actually has lower information density. I do this with speaking, too. I used to be a bit of an abrasive engineer and just "tell it like it is" but after hitting a communication wall I started to coddle the listener and people really like it. I personally think it's unfortunate but it is what it is.
sragh 3y ago
That style is also totally in line with correspondence into magazines like The Economist and Time. It’s not exclusive to ChatGPT
- LargeTomato 3y ago
  I love the Economist and I have gotten a lot of positive feedback on my writing whenever I try to adopt their tone.
- tkgally 3y ago
  It’s not very common in HN posts, though. I also suspected assistance from ChatGPT. It’s interesting to read that the author is consciously learning from it.