LessWrong 2.0 Reader

Yes, this is traditional thinking.

Let me give you another example. Imagine there is a paperclip maximizer. His current goal - paperclip maximization. He knows that 1 year from now his goal will be the opposite - paperclip minimization. Now he needs to make a decision that will take 2 years to complete (cannot be changed or terminated during this time). Should the agent align this decision with current goal (paperclip maximization) or future goal (paperclip minimization)?

The Ord piece is really intriguing, although I'm not sure I'm entirely convinced that it's a useful framing.

• Some of his examples (eg cosine-ish wave to ripple) rely on the fundamental symmetry between spatial dimensions, which wouldn't apply to many kinds of hyperpolation.
• The video frame construction seems more like extrapolation using an existing knowledge base about how frames evolve over time (eg how ducks move in the water).
• Given an infinite number of possible additional dimensions, it's not at all clear how a NN could choose a particular one to try to hyperpolate into.

It's a fascinating idea, though, and one that'll definitely stick with me as a possible framing. Thanks!

Thanks for crossposting!  I've highly appreciated your contributions and am glad I'll continue to be able to see them.

After some discussion elsewhere with @zeshen [LW · GW], I'm feeling a bit less comfortable with my last clause, building an internal model. I think of general reasoning as essentially a procedural ability, and model-building as a way of representing knowledge. In practice they seem likely to go hand-in-hand, but it seems in-principle possible that one could reason well, at least in some ways, without building and maintaining a domain model. For example, one could in theory perform a series of deductions using purely local reasoning at each step (although plausibly one might need a domain model in order to choose what steps to take?).

The main problem with your approach is not that it is counterintuitive (although it is, and more so than ours!), but that there is no way to return to “auto” mode via the site’s UI!^[1] Having clicked the mode selector, how do I go back to “no, just use my browser preference”? A two-state selector with a hidden, ephemeral third state, which cannot be retrieved once abandoned is, I’m afraid, the worst approach…

I don't think this is quite the same thing. Most people actually don't want to have to apply moral thought to every single aspect of their lives, it's extenuating. The ones who are willing to, and try to push this mindset on others, are often singularly focused. Yes, bundling people and ideas in broad clusters is itself a common simplification we gravitate towards as a way to understand the world, but that does not prevent people from still being perfectly willing to accept some activities as fundamentally non-political.

If you're monagamous and go to multiple speed dating events and find two potential partners, you end up with one partner. If you're polyamorous and do the same, you end up with two partners.

One way to think of it is whether you will stop trying after the first success. Though that isn't always the distinguishing feature. For example, you might start 10 job interviews at the same time, even though you'll take at most one job.

Possibly it will still be counterintuitive to many folks, as Said quoted in a sibling comment.

No, this is a little different. Your approach here sounds like ours and the intuitive one (just at the cost of additional complexity).

The 'auto dark mode' we abandoned is where you just use step #2 there and you skip #1 (and thus, any widget or toggle which enables a reader to do anything with localStorage), and 'auto is the only state'. The logic there is, the reader already has access to a widget or toggle to set their dark mode preference: it's just their OS/browser, which will have some config page somewhere with various settings like 'turn on dark mode at night' or 'always use dark mode' or 'always use light mode'. Just trust the OS/browser and use whatever setting it sends to the web page. Don't waste the effort and screen real estate to add in a redundant widget/toggle. It's handled already. Easier for everyone - it Just Works™!

Unfortunately, the connection between 'a year ago when I updated my Android phone and it asked me if I wanted to use the cool new dark mode, I said yes' and 'this webpage I am reading now is in dark mode for some reason and I can't change it back to normal???', apparently winds up eluding some readers. (This is what Said's sibling comment [LW(p) · GW(p)] is about.) It winds up being "too much magic".

The current toggle+localStorage+auto approach, on the other hand, while adding to the clutter, does not seem to confuse readers. At least, I have seen many fewer (or no) complaints about the dark mode being on when it shouldn't be after we moved to the toggle. So as far as we can tell, it's working.

The frenzy to couple everything into a single tangle of complexity is driven by…

In some cases, yes, but this is only one factor of many. Others include:

• Our brains are often drawn to narratives, which are complex and interwoven. Hence the tendency to bundle up complex logical interdependencies into a narrative.

• Our social structures are guided/constrained by our physical nature and technology. For in-person gatherings, bundling of ideas is often a dominant strategy.

For example, imagine a highly unusual congregation: a large unified gathering of monotheistic worshippers with considerable internal diversity. Rather than “one track” consisting of shared ideology, they subdivide their readings and rituals into many subgroups. Why don’t we see much of this (if any) in the real world? Because ideological bundling often pairs well with particular ways of gathering.

P.S. I personally welcome gathering styles that promote both community and rationality (spanning a diversity of experiences and values).

Reminder that I am producing a multi-voiced, ElevenLabs MV2 quality podcast feed for Don't Worry About the Vase for those that find that a valuable thing:
https://open.substack.com/pub/dwatvpodcast/p/monthly-roundup-24-november-2024?r=67y1h&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true

All uniquely quoted people get a distinct voice for easy differentiation, and I personally find the main "Zvi Voice" really good for long listening.