Note: other pages may also contain ratings according to PGNL. Those might conflict with the ones represented here. But I’ve timestamped when I judged the content, so thinking “when was this evaluated” might help resolve any conflicts.
| entity | precise | general | novel | live | date |
|---|---|---|---|---|---|
| The Beginning of Infinity | 0.99 | 0.99 | 0.90 | 0.8 | 2023-03-27 |
| Antifragile | 0.90 | 0.90 | 0.95 | 0.95 | 2023-03-27 |
| The War of Art | 0.70 | 0.50 | 0.90 | 0.99 | 2023-03-27 |
| Skunk Works: A Personal Memoir | 0.90 | 0.60 | 0.60 | 0.95 | 2023-03-27 |
Note: there’s no “goodness” column. “Goodness” is hard to compare. “Goodness” is vague, and intensely subject to the interpretation of the person. “Goodness” is easy to vary (thanks, David Deutsch. Also, thanks Rich Hickey for The Value of Values, ie “value judgments should be timestamped with date observed”).
If “Goodness” leads to imprecise judgment, what’s a better way? I don’t think we can reduce content to a single number. I think a better approach is by describing aspects of the content.
I’d say (precision, generality, novelty, liveness, date)
is harder to vary than (goodness).
Originally, this section was the whole article! As of 2023-03-27, I’m pushing it down in significance. Start with a table, explore theory later.
| vague | precise |
| specific | general |
| well-known | novel |
Forms a 3D space.
Pysics: precise, general, well-known.
My engineering experience: precise, quite specific, novel.
Web3 wild speculation: vague, general, novel.
Idea — always better to move to the right. But hard.
| vague | precise |
| specific | general |
| well-known | novel |
| objective | subjective |
Copied over from: https://subcons.teod.eu/specific-precise/
It’s possible to be precise about generalities (good philosophy), and it’s possible to be precise about specifics (good measurement, good planning).
For example:
| Specific | General | |
|---|---|---|
| Precise | Good measurement, solid observation, good planning | Good philosophy |
| Vague | Stuff that’s plain wrong. | Getting lost in abstraction |
Or:
| Specific | General | |
|---|---|---|
| Precise | Detail | Good logic, good mathematics |
| Vague | Stuff that’s plain wrong. | Equivocation |
Which set of examples would you prefer? Why?
Easier to draw.
I often accept a small loss of precision to gain generality. Others are often happy with me when I strive to be specific.
Here’s a table:
| axis | when 0 % | when 100 % |
|---|---|---|
| :generality | event | physical law |
| :novelty | known by all | known by 1 |
| :precision | dream | logic |
| :subjectivity | event | taste |
Then I can query on stuff.
Can either be a 0-1 scale, or :low, :medium, :high
| :low | ~0.2 |
| :medium | ~0.5 |
| :high | ~0.8 |
Default values (humble)
| :generality | :low |
| :novelty | :low |
| :precision | :low |
| :subjectivity | :high |
By default, “just saying stuff” does:
| :generality | 0.8 |
| :novelty | 0.8 |
| :precision | 1 |
| :subjectivity | 0 |
That means I can actually order by “good stuff” 🤔
good := correlation_mapT * article_things
Hmm.
Here are some contrasting views. Please don’t mind me shrugging in the middle.
| specific | Keep it specific | 🤷 | I like general theories | general |
| novel | Gimme new ideas | 🤷 | I read old knowlege | time tested |
| precise | Use words right | 🤷 | the journey is to dream ❤️ | poetic |
| data | Give me facts. | 🤷 | What do you like? | emotion |
I can make a nice little UI that works on this. Set your preference. Load preferences from others.
See: https://dorian.substack.com/p/at-any-given-moment-in-a-process
What is liveness? What is degree of life?
Definitions:
Just rated Antifragile and The Beginning of Infinity.