Tyler Mills’s avatar

Tyler Mills

@tyler-mills​·​Joined Jan 2026​·​Ideas
Log in or sign up to follow Tyler or post on their wall.
  Tyler Mills commented on criticism #4827.

Reiterating/refining #3904: I think the yellow "Criticism of" bubbles can and should be replaced by a graphical indication that is much easier on the eyes. The dropdown line can be made red if the comment it leads to is a criticism, and the bubble on the criticism can be eliminated. Reading the yellow bubble to get the idea # it is referring to, then searching the ideas above for the matching # is inelegant (even if it is usually the one right above).

#4827​·​Tyler Mills, 24 days ago

And dimmer red for refuted criticisms, brighter red for pending ones! Default gray for comments.

  Tyler Mills addressed criticism #4828.

The yellow bubbles link to the ideas they are criticizing, which can be handy.

#4828​·​Tyler Mills, 24 days ago

The link could be put in a new tooltip, or something. Or kept as is, just without the yellow bubble, frankly.

  Tyler Mills addressed criticism #4830.

The quote

indentation bar

is red, which would cause visual confusion.

#4830​·​Tyler Mills, 24 days ago

It should be made not red. Gray. Arguable even without the red criticism line idea above, since it already conflicts with the "red = criticism" motif.

  Tyler Mills addressed criticism #4827.

Reiterating/refining #3904: I think the yellow "Criticism of" bubbles can and should be replaced by a graphical indication that is much easier on the eyes. The dropdown line can be made red if the comment it leads to is a criticism, and the bubble on the criticism can be eliminated. Reading the yellow bubble to get the idea # it is referring to, then searching the ideas above for the matching # is inelegant (even if it is usually the one right above).

#4827​·​Tyler Mills, 24 days ago

The quote

indentation bar

is red, which would cause visual confusion.

  Tyler Mills addressed criticism #4828.

The yellow bubbles link to the ideas they are criticizing, which can be handy.

#4828​·​Tyler Mills, 24 days ago

Is it handy? I have yet to want to open the criticized idea in a new tab. I have only ever wanted to scroll up to see it, which is slightly irksome with the current yellow bubble hashtag-matching method. And when the criticized idea is clearly immediately above, the yellow bubbles serve no real purpose, only add visual clutter.

  Tyler Mills addressed criticism #4827.

Reiterating/refining #3904: I think the yellow "Criticism of" bubbles can and should be replaced by a graphical indication that is much easier on the eyes. The dropdown line can be made red if the comment it leads to is a criticism, and the bubble on the criticism can be eliminated. Reading the yellow bubble to get the idea # it is referring to, then searching the ideas above for the matching # is inelegant (even if it is usually the one right above).

#4827​·​Tyler Mills, 24 days ago

The yellow bubbles link to the ideas they are criticizing, which can be handy.

  Tyler Mills posted criticism #4827.

Reiterating/refining #3904: I think the yellow "Criticism of" bubbles can and should be replaced by a graphical indication that is much easier on the eyes. The dropdown line can be made red if the comment it leads to is a criticism, and the bubble on the criticism can be eliminated. Reading the yellow bubble to get the idea # it is referring to, then searching the ideas above for the matching # is inelegant (even if it is usually the one right above).

  Tyler Mills addressed criticism #4616.

Not sure this is a good idea. You say you wouldn’t mind horizontal scrolling, but users generally dislike horizontal scroll.

#4616​·​Dennis HackethalOP, about 2 months ago

Could be optional, as I said. Rearrange top-level ideas as toggled. Maybe not worth the trouble. Just spitballing. See #4825.

  Tyler Mills addressed criticism #4617.

Unclear how comments would be rendered.

#4617​·​Dennis HackethalOP, about 2 months ago

Not understanding this criticism. Maybe my idea is unclear. I'm picturing the existing "column" of a discussion, repeated column-wise for each top-level idea. Current discussion content takes up only the left ~third of my screen, while the right two thirds of my screen is totally unused. The cost of using that real estate is more content (clutter) on screen, the benefit is less time scrolling up and down in one dimension, looking for given ideas and getting bearings, which I sometimes find tiring. A second dimension helps get bearings (e.g. "Oh yeah, this relates to that one over here near the middle of the third column." Rather than: "That one was ... 77% of the way down the page, hmm, what were some words from it that I can use to ctrl+f, grrrrr.").

  Tyler Mills posted idea #4824.

Thoughts on an optional "implies" relation for ideas? I find myself commenting on one idea something which it implies, then criticizing that, but the original idea is not marked criticized. Being able to chain or bundle ideas avoids the bookkeeping issue of having to make new criticisms for each step in the chain, if one is criticized.

  Tyler Mills commented on criticism #4813.

Creativity isn't defined by its outputs but by its process. RNGs do not recognise or criticise ideas.

#4813​·​Dirk Meulenbelt, 25 days ago

Agreed on both counts, but I think the bountied idea survives this...
Recognizing and criticizing ideas could be a requisite for tractably synthesizing any possible explanation (I suspect as much).

  Tyler Mills commented on criticism #4816.

Speed is a property of programs, too. https://en.wikipedia.org/wiki/Big_O_notation

#4816​·​Dennis Hackethal, 25 days ago

Ah, so if I understand correctly, there are two knobs affecting speed (elapsed time) for a given algorithm: the hardware, and the implementation of the algorithm. The given algorithm has a complexity, independent of those two, which is how the time and memory scales with an input.

  Tyler Mills revised idea #4740.

Assumption A1: Only programs that are people while running constitute qualia/experience/subjectivity/consciousness.

Assumption A1: Only programs that are people, while running, can constitute qualia/experience/subjectivity/consciousness.

  Tyler Mills commented on idea #4764.

In everyday English, we say ‘probably’ to leave room for error and communicate some uncertainty. That’s fine because everyone knows we’re not assigning actual probabilities in the sense of the probability calculus.

In math, we use the probability calculus to describe the frequency of outcomes for underlying processes that look random. Like a coin toss. That’s also fine because we know all possible outcomes and we have a measure for each.

Things go wrong when people use probability even though they don’t know the outcomes (because of the growth of knowledge, say, as you write in #4762) or they have no measure for them or the underlying phenomena don’t behave randomly (again because of the growth of knowledge). Like Elon Musk tweeting we’re 90% likely to see AGI in 2026. (Not a literal quote but he says stuff like that sometimes.)

Some people try to steal the prestige of math and hide their ignorance by using the probability calculus illegitimately.

See also https://www.youtube.com/watch?v=wfzSE4Hoxbc. It’s been years since I watched it but it’s bound to have related ideas.

#4764​·​Dennis Hackethal, about 1 month ago

I think this clarifies it, thanks. Used in common speech to express uncertainty and leave room for error, valid mathematically only when all outcomes are known and have a measure.

  Tyler Mills commented on idea #4760.

In summer in the desert, will it "probably" be sunny in the afternoon?

#4760​·​Tyler MillsOP, about 1 month ago

No, it will or it won't, but "probably" expresses one's awareness of a lack of battletested explanations, or of their own uncertainty or lack of confidence in the prediction, etc. (cf. #4764).

  Tyler Mills commented on criticism #4807.

Doesn't it? All explanatory knowledge is in the set of all possible programs, and a random program (or number) generator can generate any of those, given infinite time.

#4807​·​Tyler MillsOP, 25 days ago

We could say a person is a program that can synthesize any possible explanation in finite time, excluding memory limitations. But this would again grant personhood to RNGs. For that matter, a counting program could just enumerate all possible binary strings up to its memory limit, in finite time...

  Tyler Mills addressed criticism #4783.

Understanding explanatory knowledge seems like a better criterion

#4783​·​Knut Sondre Sæbø revised 27 days ago

Maybe... but "understanding" is too vague, I think. Doesn't understanding mean: can explain? But then this is just "can create any explanation" again. I think the core question is why a random program generator isn't a person, coming from Deutsch's definition of a person as a program that has explanatory universality -- can create any explanation (my thought here is that this definition isn't good enough on its own, given the random generator point).

  Tyler Mills addressed criticism #4781.

A random number generator does not create explanatory knowledge.

#4781​·​Dirk Meulenbelt, 28 days ago

Doesn't it? All explanatory knowledge is in the set of all possible programs, and a random program (or number) generator can generate any of those, given infinite time.

  Tyler Mills addressed criticism #4805.

But an AI programmed to make random variations to its conjectures (English or otherwise) can only do so by choosing from an existing set of variations. Again, that knowledge is pre-existing. True evolution must involve variations to the substrate on which the knowledge is based; variations must be agnostic to the semantics of whatever they are acting upon, else they are already implicit from it, in which case their application does not constitute a truly novel conjecture (in the sense defining creativity).

#4805​·​Tyler MillsOP, 27 days ago

Even if variations are agnostic to any meaning or context of the knowledge, why are they still not implicit? Anything is implicit from anything else, if implicit just means: follows from when a given change is applied... The whole question is where the change is coming from... (?)

  Tyler Mills addressed criticism #4803.

If only some of the criteria are stored, and the rest are random, is it still evolution? Is evolution only happening if there is random variation? But we could program an LLM to do that as well...

#4803​·​Tyler MillsOP revised 27 days ago

But an AI programmed to make random variations to its conjectures (English or otherwise) can only do so by choosing from an existing set of variations. Again, that knowledge is pre-existing. True evolution must involve variations to the substrate on which the knowledge is based; variations must be agnostic to the semantics of whatever they are acting upon, else they are already implicit from it, in which case their application does not constitute a truly novel conjecture (in the sense defining creativity).

  Tyler Mills revised criticism #4801.

If only some of the criteria are stored, is it still evolution? Then evolution is only the random part of the variation?

If only some of the criteria are stored, and the rest are random, is it still evolution? Is evolution only happening if there is random variation? But we could program an LLM to do that as well...

  Tyler Mills revised idea #4800 and marked it as a criticism.

If only some of the criteria are stored, is it still evolution? Then evolution is only the random part of the variation?

If only some of the criteria are stored, is it still evolution? Then evolution is only the random part of the variation?

  Tyler Mills commented on criticism #4799.

Whatever new "explanations" it creates are derivable from (and by?) the knowledge in the training data. It isn't evolution if all of the variations and selection criteria are stored ahead of time. That's just a search process, as in the case of Move 37 per AlphaGo.

#4799​·​Tyler MillsOP, 27 days ago

If only some of the criteria are stored, is it still evolution? Then evolution is only the random part of the variation?

  Tyler Mills addressed criticism #4797.

only people can create explanatory knowledge

How is an LLM not creating new explanatory knowledge (even if worse than the existing, by any measure), by varying some existing written explanation? It could even vary and select by some criterion of its "choice", thus realizing Popperian epistemology.

#4797​·​Tyler MillsOP, 27 days ago

Whatever new "explanations" it creates are derivable from (and by?) the knowledge in the training data. It isn't evolution if all of the variations and selection criteria are stored ahead of time. That's just a search process, as in the case of Move 37 per AlphaGo.

  Tyler Mills criticized idea #4722.

The definition of fitness that rendered Move 37 the best choice originated outside the system.

#4722​·​Tyler MillsOP, about 1 month ago

The definition of fitness for DNA also originated outside it, so this doesn't in itself suggest the system isn't actually creating new knowledge.