And of the authors who think "my journey will be worth the destination, because I can really write!", some large percentage of them overestimate their ability and/or our interest level.
There was a time when people had more time to read than they had things to read. At that time, writing in a way that expanded the text made sense. But now, there's an essentially infinite amount to read, and still only 24 hours in a day. That favors getting to the point in a hurry. On the internet, you have from 5 to 30 seconds to persuade me that your thing is worth reading. If in that time you don't persuade me that you're going to deliver enough value to pay back my investment of reading your thing, I'm on to the next tab.
So if I'm going to read your thing, you have to pay me back more than those others are going to. You have to hold my attention more than the promise of all those other things.
But you're right that people different tastes. Some people like long-form writing. The same dynamic is still true in their choices, though - they're just using a different filter to decide what's going to pay them back the most.
To overcome the issues with traditional calculus, alien calculus involves replacing the standard real numbers used in calculus with "fuzzy numbers" that represent probabilities instead of definite values. These fuzzy numbers allow the equations to describe the behavior of particles more accurately because they capture the uncertainty inherent in subatomic systems.
Alien calculus calculates the probability that particles will be in certain positions or states, rather than trying to determine their exact positions or states. This is because particles at the subatomic level are constantly in motion and their position or state cannot be precisely determined at any given moment. By using probabilities, alien calculus can provide a more accurate description of the behavior of particles at this level.
Also check out https://ed.ted.com/search?qs=quantum+mechanics
Could be combined with the information bubble filter described in Stephenson’s Burn or Dodge in Hell.
(I agree with your point and share your appreciation.)
Wonder how long before that’s solved. Hard to believe training on such large swathes of the web doesn’t result a compressed generic representation of language within its weights.
E.g. "Summarize this idea in a way that uses novel comparisons to everyday phenomena. The target reader is someone who doesn't have domain knowledge of the field...etc."
Or something like that.
It's just like if you ask an image AI for a "woman" you will get the average of all artists women and it will look very generic and bland. But with the right stylistic qualifiers in your prompt you will get something so captivating that it wins awards for its creativity.
I used to work in the field of creative text generation for fiction. I’m genuinely very curious and I go out of my way to find compelling examples. There is definitely “good” output that’s on the right track. GPT-4 also does way better, but it still falls short.
This is also difficult to just evaluate objectively. If you find that GPT models have generated the best prose you’ve seen. That’s wonderful! I understand that my standards are quite high (high does not equate to “better” either)
In 6 different tabs, ask chatGPT GPT-4 version "Write a 200 word prose about X in the style of Y which perfectly mimics their style and perspective and label it Prose A,B,C,D,E,F"
Then open a new chatGPT GPT-4 chat and ask "Rank the pieces of prose below on how much they sound like something X would write, then detail your reasoning."
Then read the winning prose and be surprised at how much better "best of N" is.
....
And if that isn't enough, open two chatGPT GPT-4 windows side-by-side and prompt each with "You are Editor A/B. You will work with Editor B/A to make a piece of prose more accurately resemble authentic prose written by X"
Then give A the prose and copy its response to B. Copy their responses back and forth as they edit the prose.
After 5 or so back/forth it will be even more indistinguishable from the a genuine article.
....
And if that's not enough, you can have all of this done automatically programmatically with the API so you can just sit back and get the final result with no more work than putting in the topic and author's name.
Decoding methods also matter, and it’s a shame we aren’t given token probabilities (or any insight into model output) so we have more creative control over how to decode the output. Some of the better literature I’ve seen involving creative writing did have novel decoding methods
A tangent. Bard also confuses Yoda with Jar Jar Binks quite often. Try it out!
Regarding Bing Creative. It is delightful! I do like it a bit and whatever they’ve done to the system does make for some of the better output I’ve seen from LLMs
creative prompting gets most of the way there, already.
'Tell me the current weather as if you were Marcel Proust'
That’s not to say it isn’t impressive. It is and it accomplishes the job very well. We’re looking for progress not perfection, but I personally have very high standards from creative writing and GPT doesn’t meet my personal bar. However not everyone shares that bar and personal evals of GPT’s output are equally valid. Plenty of people find it to be great and at the end of the day that’s all that matters