Nice work!

ChatGPT on the other hand was able to reformulate the explanation until I understood the part I was struggling with, namely what prevents a proxy from simply passing its own public key as the website's public key. It did so without citing anything.

I did this using cursor. You can use other methods[1], but I'll describe mine.

1. Grab the text contents of a style guide you like (I used Strunk, which is easily available on internet. Use whatever you like.)

2. Paste that into Cursor as a txt file.

3. Paste your target text into Cursor as another txt file.

4. In the AI Pane, switch your model to Gemini 3 Pro (for its intelligence + long context)

5. Prompt it thusly:

""""""""""""

Can you please create a paragraph-for-paragraph rewrite of NAME-OF-TARGET-TEXT using NAME-OF-YOUR-STYLE-GUIDE?

All the original ideas must be maintained -- we are simply trying to create a version that matches the style guide better. The paragraph ordering must be matched.

Please output this in OUTPUT-FILE.txt

Before doing so, I want you to create a todo list (it will be quite long) of the form:

- Convert Book X Part Y

- Convert Book X Part Z

And so on, exhaustively covering all parts.

You will have to view the entire file to do this.

You must ALWAYS output a new part after the old part, never overwrite.

You must always convert a part, output it, and then move on.

Your flow should be:

(1) view NAME-OF-YOUR-STYLE-GUIDE (if they are no longer visible)

(2) view part of NAME-OF-TARGET-TEXT you're converting

(3) output converted part into OUTPUT-FILE.txt

(4) check it off on your todo list

(5) begin next part

""""""""""""

[1] Though I used cursor, you should be able to script the whole "paragraph for paragraph" thing if you wanted to, which I might do later. (If you do this, let me know!)

output = layers(layers(layers(layers(input))))

instead of the classical:

output = layer4(layer3(layer2(layer1(input))))

output = layers(input)

Or

output = layers(layers(input))

Depends on how difficult the token is.

    x = tokenize(input)
    i = 0
    do {
      finish, x = layers(x)
    } while(!finish && i++ < t_max);
    output = lm_head(x)

    x = tokenize(input)
    i = 0
    finish = 0
    do {
      p, x = layers(x)
      finish += p
    } while(finish < 0.95 && i++ < t_max);
    output = lm_head(x)

as a nearly perfectly uninformed occasional consumer of space-related articles, I can say it makes a lot of sense (to me) why we'd use something like that to move things around the solar system.

Sharing the funniest Christmas themed video I know