Simple but important ideas
Important ideas don’t always require long explanations. Here’s a famous example:
Let an ultraintelligent machine be defined as a machine that can far surpass all the intellectual activities of any man however clever. Since the design of machines is one of these intellectual activities, an ultraintelligent machine could design even better machines; there would then unquestionably be an “intelligence explosion,” and the intelligence of man would be left far behind… Thus the first ultraintelligent machine is the last invention that man need ever make.
This is a single paragraph on the third page of a 50 page report. Maybe someone who’s good at parsing 60s era academic English can tell us if the rest is any good.
It seems like anyone who has an idea they want people to take seriously has to write a bunch about it. This is most apparent in popular nonfiction books, which are often bloated far beyond what it takes to communicate the core ideas.
To correct this “presentation length bias”, we can fight it from both ends:
Remember that important ideas don’t have to be in an important place, be said by an important person, or be an important length.
Alert readers to important ideas that don’t look important (e.g. “This is a simple idea, but it seems important:”). Do this especially if it’s someone else’s idea, since people are going to be reluctant to label their own ideas as important.
Somewhat related:
The Center for Communicating Science, together with Alan Alda, is sponsoring what they call the Flame Challenge: Answer the question – “What is a flame?” – in a way that an 11-year-old would find intelligible “and maybe even fun.”
The answer “oxidation” has the virtues of being both short and strictly correct, but, given the audience, missed the point entirely. I assume the answer that the sponsors of this contest are looking for will also be rather short and strictly correct as well, although longer than a one-word label.
In my current mood I would guess that non explanations such as “oxidation” are to some extent are responsible for the Romantic reaction against the scientific endeavor. It can be hard to take joy in the merely real if those explaining the merely real aren’t very good at it.
SEO tip: write a lot about your idea. Then searchers will stumble upon it using lots of different keywords. Ideally, outsource this job to unpaid commentators and critics.
One thought: writing a lot about an idea is an indicator that the person sharing it has thought about it a lot.
Related thought: More details, even unimportant ones, can act as a near-mode anchor.
Length imparts status. The size of swords in japanese games and anime should illustrate this. :p
“The size of swords...”
Long posts → compensating for other shortcomings? :)
“If I had more time, I would have written a shorter letter.”—Blaise Pascal (and probably some other people too)
A noteworthy counter-point is that not all important ideas are simple.
Or simple ideas have all sorts of implications that naturally follow, but which most readers need to have teased out for them.
Fermat’s last Theorem: There is no positive integers x, y, z, and n > 2 such that x^n+ y^n = z^n.
Important, yes. But it took three centuries and far more than a single line of algebra to prove the idea. An idea can be succinct, whereas a proof (whether formal or casual) is generally going to be far more complex. Ideas without proof are only useful when you’re willing to trust an appeal to authority, or you’ve already verified the proof to sufficient confidence for the task at hand.
Objectivism can be summarized in four sentences. In this four-sentence form, it’s not extremely compelling. In the form of thousands of pages of fiction, however, it becomes very seductive and has in fact become an “important idea” as far as history is concerned, regardless of its actual merit.
Communication is hard, and even good communicators are unlikely to successfully convey 100% of their intended message to 100% of their target audience on the first try. So even good communicators, if they have what they think is an important idea, will be tempted to communicate that idea again and again and again. The Sequences are over a million words long, and a lot of those words are intentionally repetitive, calculated callbacks to previous Sequence posts establishing a connectome of ideas.
That said, I agree that the core, most minimal expression of an idea or argument is valuable and probably a good exercise to undertake if you are trying to teach something.
I’m reminded of the story of the gentile asking the rabbi Hillel that the whole Torah be explained to him while he stood on one foot. Hillel replied “What is hateful to you, do not do to your fellow: this is the whole Torah; the rest is commentary; go and learn.”
An article in the Jewish Daily Forward observes that it is a mistake to equate “the rest is commentary” with “the rest is unimportant.”
This post is an example of a simple but important idea.
I certainly have the same tendency. Whenever I come up with an insight I consider important, I always find myself trying to make sure the presentation is long enough (whatever the hell that means). Just three sentences? Make it two paragraphs! Just 500 words? What about 2,000!
As you said, we see this in the book publishing business, where with no exaggeration nearly every non-fiction book seems at least an order of magnitude longer than what would be necessary and efficient for raw communication, but we can allude to the explanation of this fact with a simple observation: it’s difficult to publish a 20-page manuscript and sell it for $20 each.
But why does the presentation length bias you speak of seem to apply so broadly? Not just to books, but to everything. So often when somebody has what they think is an important idea, they find it necessary to write a useless, wandering introduction; pepper their presentation with point after point that sounds related but doesn’t add anything to the real insight to be had; and so on. Anything to make it longer, dammit!
We must combat this tendency. As a community, we should try to establish norms that reward explaining important points concisely, and acknowledge that when the goal is just communication (rather than signaling), length is a cost. We should do what we can to optimize for a community with the highest reading investment to information return as possible, the best signal to noise ratio manageable.
I should mention though that the sheer character or word count of an article or comment is only one factor in its brevity. One comment may be twice as long as another in terms of its word count, but be twice as readable and easy to get through. Longer sentences can be more concise than shorter ones if they’re written to flow really well and optimized for quick assimilation.
There’s potential for lost purpose here. The character count is certainly a strong indicator for how concise it is, but it’s not a perfect one. Just because you can remove five words from a sentence and have it still convey the same meaning doesn’t necessarily suggest that you should do so. It may end up flowing worse, or being harder to follow because of some oddity about our language hardware, etc.
Anyway, if any community is in a position to do away with this bias, or at least make it much less pervasive, it’s this one. Length is a cost, or rather a high time and effort requirement is a cost. I’m here to learn, and do so as quickly and efficiently as possible. If I can get the same amount of important information in 1⁄10 of the time, that’s great! Whatever signaling function bloating the length carries, we must find a substitute (if possible)!
Summary: Length is a cost.
Plus a bunch of other (perhaps useful and relevant) details. Either that, or I just bloated it well past what would have been necessary, and did so solely for the purpose of making it look important.
Is there a way, to make this last needed invention far simpler than anybody here dreams about?
An insight, how to build it astonishingly simple, just as a super-intelligence would?
Interesting. What new insights can we get from asking ourselves how an AI would build an AI?
A super-intelligence would constantly ponder what is the next optimal action and then it would execute this calculated action.
I try to operate that way, for a while now.
Nevermind