Comment author: whpearson 02 June 2009 01:42:05PM 0 points [-]

I'm trying to argue more than that, does my edit to the post make it clearer?

Comment author: JulianMorrison 02 June 2009 03:07:03PM -1 points [-]

A bit. You're arguing that one intelligent system, comprising smarts and knowledge, can be differently effective depending on its context. Smarts might be less effective if opposed by something smarter. Knowledge might be less effective if it's mistaken or incomplete.

So far, not controversial.

What you haven't managed to do is dent the recursive self improvement hypothesis. That is, you haven't shown that "all things aren't equal" between an AI and its improved descendant self.

Comment author: JulianMorrison 02 June 2009 12:07:13PM -1 points [-]

You seem to be arguing that one FOOM in the middle of another bunch of faster FOOMs won't be so impressive.

Er, OK, fair enough, and so?

Comment author: MrHen 15 May 2009 02:35:00AM 0 points [-]

Is it valid when considering kidnapping?

In response to comment by MrHen on Closet survey #1
Comment author: JulianMorrison 15 May 2009 09:19:54AM 0 points [-]

Didn't we already beat that one to death? The child's volition isn't all that's involved with kidnapping. It isn't directly comparable.

In response to comment by Alicorn on Closet survey #1
Comment author: MrHen 15 May 2009 12:13:44AM 1 point [-]

Ooh, yes, you are very right. Apologies.

In response to comment by MrHen on Closet survey #1
Comment author: JulianMorrison 15 May 2009 01:05:18AM 1 point [-]

OK, so, we'll go with entice.

Enticing would usually mean suggesting the activity is intrinsically desirable, offering a trade, asking pretty please, making a dare, or etc. We'll assume the child's mind is changed by the enticement.

Why would that change not simply be valid?

Comment author: MrHen 14 May 2009 05:59:21PM 0 points [-]

Removing a child from a parent is a harm (as witness the panicked parent). It's not so much a matter of consent, as of making people worry and separating them from their family.

If the child is returned before the parent knows they are missing? I am not understanding why the correlation is so hard to see. It is an analogy, not a mirrored situation. Kidnapping is not seducing. There are differences. The original point was that seduction involves coercing children. Kidnapping can do the same thing. So can brainwashing. All three of these (kidnapping, brainwashing, seducing) can produce harm but may not and arguing about exactly when "harm" happens is not really useful. The relevant question is exactly this:

BTW, this line of argument doesn't get you to "no sex", it gets you to "no sex without parental consent". Fair enough, now what if they say "yes"?

I am not arguing for any particular stance. I just saw an interesting correlation between seduction and kidnapping that involved coercion. If I remember correctly, the laws in some states get remarkably relaxed when minors have their parents' consent. I could not tell you specifics, however. If you find this sort of thing interesting I am sure it is relatively easy to find information about sex with parental consent.

The bottom line: A child will do an awful lot to please someone. Is it okay to coerce them into doing something? Does it matter if they enjoy it? Does it matter if there is harm? Does it matter if they want to do it?

All of this also assumes "seduction" instead of a real, true romance. I would assume that a real, true romance has less coercion. (Or, at the very least, thinks it has less coercion.)

In response to comment by MrHen on Closet survey #1
Comment author: JulianMorrison 14 May 2009 10:12:22PM 2 points [-]

Perhaps we're being confused by your use of the verb "seduce", since to me that doesn't include non-consensual means - it usually implies cunning trickery at worst and goal-directed charm at best. Can you restate without using it?

Comment author: MrHen 14 May 2009 02:30:08PM *  1 point [-]

I always assumed that part of the problem is that it is easier to coerce children. If I kidnap a child and do nothing but feed them ice-cream and take them on a tour of the zoo it is still wrong, even if they liked it and no harm was done.

No ransom and not against the child's will. If the reason kidnapping is wrong deals with parental consent, does the same thing apply to sex?

But what are the separate harms of sex?

This is actually irrelevant for the point I was trying to make. Kidnapping, with no harm done, is still very much illegal. Should it be?

In response to comment by MrHen on Closet survey #1
Comment author: JulianMorrison 14 May 2009 02:53:20PM 6 points [-]

Removing a child from a parent is a harm (as witness the panicked parent). It's not so much a matter of consent, as of making people worry and separating them from their family. The parents have a protective interest in the child, which is harmed by their non-consent to the zoo trip. This is the very thing that makes it "kidnapping" and not "visiting with friends". It is a separate harm, which is why the distinction I drew is relevant.

BTW, this line of argument doesn't get you to "no sex", it gets you to "no sex without parental consent". Fair enough, now what if they say "yes"?

Comment author: MrHen 12 May 2009 09:45:22PM *  4 points [-]

(Edit) During this entire thread I was misusing the word "coerce." I meant something more like "entice." Thanks Alicorn.

If as it turns out, kids enjoy consensual sex and take no harm by it, on what basis can society consider it wrong? There has to be a reason. Societies can't just create moral crimes by their say-so.

I always assumed that part of the problem is that it is easier to coerce children. If I kidnap a child and do nothing but feed them ice-cream and take them on a tour of the zoo it is still wrong, even if they liked it and no harm was done.

If I seduce a child and do nothing but feed them ice-cream and have sex with them... is it still wrong? Even if they liked it and no harm was done? There are certainly risks involved and assuming things will be okay is naive. But is assuming things will be bad/evil/gross just as naive?

Suppressing the moral gag reflex is hard to do. I do not know if I can answer the question objectively. I know if I had kids I do not want anyone coercing them into having sex.

In response to comment by MrHen on Closet survey #1
Comment author: JulianMorrison 14 May 2009 10:09:29AM 5 points [-]

If I kidnap a child [...]it is still wrong

Well yes, because kidnapping involves taking a child from their parents unannounced, possibly against the child's will too, possibly also asking for ransom, etc. Those are separate harms that happen even if the child enjoyed the ice-cream and the trip to the zoo.

But what are the separate harms of sex? There are health risks, but they don't hugely exceed the risks in other common childhood activities such as tree climbing.

Comment author: JulianMorrison 14 May 2009 01:25:12AM 5 points [-]

Original Naziism was very tied to a historical milieu. The conditions of Weimar Germany won't exactly recur. But by interpreting it as a suite of metaphors, it becomes somewhat immune to context, more able to bridge the centuries - while continually enticing its adherents to drag the world back to 1935, which is the era in which the metaphors make sense.

Comment author: JulianMorrison 07 May 2009 10:53:15PM 0 points [-]

I find most koans non-contradictory. In this, Joshu teaches the moment of free cognition during surprise. Yes and no run in grooves of assumption. Joshu stomps the grooves flat. The mind will quickly cut a new groove, but for an instant inputs are considered as themselves, and you actually see the dog.

Comment author: gwern 07 May 2009 01:15:26AM *  19 points [-]

'I was reminded of this recently by Eliezer's Less Wrong Progress Report. He mentioned how surprised he was that so many people were posting so much stuff on Less Wrong, when very few people had ever taken advantage of Overcoming Bias' policy of accepting contributions if you emailed them to a moderator and the moderator approved. Apparently all us folk brimming with ideas for posts didn't want to deal with the aggravation.'

I don't really have a point here, but this shouldn't really be surprising at all, not at this moment in time.

I mean, has anyone here not used Wikipedia? (I'd also wager even odds that >=90% of you have edited WP at some point.)

EDIT: Looking back, it seems to me that what would not be surprising is, upon observing LW suddenly skyrocketing in contributors & contributed material, noticing that the sudden increase comes after a loosening of submission guidelines. When a site skyrockets, it's for one of a few reasons: being linked by a major site like Slashdot, for example. Loosening submission guidelines is one of those few reasons.

But that's not to say that Eliezer should have confidently expected a sudden increase just because he loosened submission criteria; the default prediction should have been that LW would continue on much as OB had been going. Lots of wikis never go anywhere, even if they let anyone edit.

Comment author: JulianMorrison 07 May 2009 03:10:11PM 3 points [-]

When you visit your friend, he says "help yourself from the kitchen". Which read literally would give you the ability to strip the kitchen bare. Obviously it doesn't mean that. If the friend had spoken as they meant, "take a reasonably small amount of drinks and munchies for immediate use, and not the fancy stuff or tonight's supper", then he would be read as being under-generous.

I suspect people run similar "what subset of their generous offer ought I to take" calculations on any wide-open offer. Taking the whole offer would be greedy.

View more: Next