There was a Pebblesorter of lore who said that all of the heaps were merely transient, that none of them would last, all eventually destroyed by increasing entropy in the universe, and that therefore none of them held any true or real satisfaction. He said that the only path to enlightenment was to build no heaps at all for to do so could only increase suffering in the world. Then the other Pebblesorters killed him.
I mean, I understand that it was the thing to do that Pebblesorters would endorse, that part isn’t startling, but I didn’t think you endorsed that “Pebblesorter::(should, right, moral, etc.)” way of speaking.
Does this reflect a change in your position, or have I misunderstood you on this all along?
Sorry, I usually do try to avoid that, but in this case I didn’t see how to form that sentence without using the word “should” because it’s traditional in “as well X should”. Keep in mind that according to C++ namespacing conventions, something inside a namespace has literally nothing to do with its meaning in any other namespace.
No apology is needed, certainly not to me; I generally treat “should” and similar words as 2-place predicates in the first place. (Well, really, N-place predicates.)
I think of them as two-place predicates, but with one of them curried by default indexically, much like in a member function in C++ foo means this->foo unless otherwise specified. (I already made that point in the second edit to this comment.)
Yeah, that makes sense as far as it goes, but I find that humans aren’t consistent about their defaulting rules. For example, if I say “X is right” to someone, there’s no particular reason to believe they’ll unpack it the way I packed it.
That can be all right if all I want to do is align myself with the X-endorsing side… it doesn’t really matter what they understand, then, as long as it’s in favor of X.
But if I want to communicate something more detailed than that, making context explicit is a good habit to get into.
Even with the disadvantage of sometimes coming across as condescending, or even often coming across as condescending to particular people, this is excellent advice.
Keep in mind that according to C++ namespacing conventions, something inside a namespace has literally nothing to do with its meaning in any other namespace.
Using this reasoning advocate a style of word usage strikes me as dubious reasoning even though the usage and real reason for using it happen to be be sensible. It screams out against my instincts for how to use words. In this kind of case if there wasn’t a clear relationship between the two functions you (hopefully) just would not even have considered using the same word.
I also note that in C++ the following also have literally nothing to do with each other, apart from the suggestive name, so C++ (and English, for that matter) are just as comfortable with “As well they should have”.
I mean, I understand that it was the thing to do that Pebblesorters would endorse, that part isn’t startling, but I didn’t think you endorsed that “Pebblesorter::(should, right, moral, etc.)” way of speaking.
It does seem to be change. In past conversations about his ‘should’ definition he has advocated ‘would-want’ for this kind of concept and carefully avoiding overloading ‘should’.
He wasn’t endorsing that position. He was saying “pebblesorters should not do so, but they pebblesorter::should do so.”
ie, “should” and “pebblesorter::should” are two different concepts. “should” appeals to that which is moral, “pebblesorter::should” appeals to that which is prime. The pebblesorters should not have killed him, but they pebblesorter::should have killed them.
Think of it this way: imagine the murdermax function that scores states/histories of reality based on how many people were murdered. Then people shouldn’t be murdered, but they murdermax::should be murdered. This is not an endorsement of doing what one murdermax::should do. Not at all. Doing the murdermax thing is bad.
He wasn’t endorsing that position. He was saying “pebblesorters should not do so, but they pebblesorter::should do so.”
You didn’t understand what TheOtherDave said. He was talking about the same usage you are talking about and commenting that it is in contrast to Eliezer’s past usage (and past advocacy of usage in conversations about how he uses should-related words.)
There was a Pebblesorter of lore who said that all of the heaps were merely transient, that none of them would last, all eventually destroyed by increasing entropy in the universe, and that therefore none of them held any true or real satisfaction. He said that the only path to enlightenment was to build no heaps at all for to do so could only increase suffering in the world. Then the other Pebblesorters killed him.
As well they Pebblesorter::should have!
I’m startled by this comment.
I mean, I understand that it was the thing to do that Pebblesorters would endorse, that part isn’t startling, but I didn’t think you endorsed that “Pebblesorter::(should, right, moral, etc.)” way of speaking.
Does this reflect a change in your position, or have I misunderstood you on this all along?
Sorry, I usually do try to avoid that, but in this case I didn’t see how to form that sentence without using the word “should” because it’s traditional in “as well X should”. Keep in mind that according to C++ namespacing conventions, something inside a namespace has literally nothing to do with its meaning in any other namespace.
I tried compiling your comment, but it didn’t work. You should adhere to the C++ conventions more closely.
You’re saying it’s a suggestively-named C++ token?
No apology is needed, certainly not to me; I generally treat “should” and similar words as 2-place predicates in the first place. (Well, really, N-place predicates.)
I was just startled and decided to ask.
I think of them as two-place predicates, but with one of them curried by default indexically, much like in a member function in C++
foomeansthis->foounless otherwise specified. (I already made that point in the second edit to this comment.)Yeah, that makes sense as far as it goes, but I find that humans aren’t consistent about their defaulting rules. For example, if I say “X is right” to someone, there’s no particular reason to believe they’ll unpack it the way I packed it.
That can be all right if all I want to do is align myself with the X-endorsing side… it doesn’t really matter what they understand, then, as long as it’s in favor of X.
But if I want to communicate something more detailed than that, making context explicit is a good habit to get into.
Even with the disadvantage of sometimes coming across as condescending, or even often coming across as condescending to particular people, this is excellent advice.
Using this reasoning advocate a style of word usage strikes me as dubious reasoning even though the usage and real reason for using it happen to be be sensible. It screams out against my instincts for how to use words. In this kind of case if there wasn’t a clear relationship between the two functions you (hopefully) just would not even have considered using the same word.
I also note that in C++ the following also have literally nothing to do with each other, apart from the suggestive name, so C++ (and English, for that matter) are just as comfortable with “As well they should have”.
Action should(Human aHumans);
Action should(PebbleSorter aPebbleGuy);
It does seem to be change. In past conversations about his ‘should’ definition he has advocated ‘would-want’ for this kind of concept and carefully avoiding overloading ‘should’.
Given his subsequent response and (I assume his) retraction of the comment, I infer that he still endorses the same position.
Yes, but clearly without being dogmatic or obsessive about it. Probably a good way to be.
What did you think his position was?
Roughly the one he articulates here.
He wasn’t endorsing that position. He was saying “pebblesorters should not do so, but they pebblesorter::should do so.”
ie, “should” and “pebblesorter::should” are two different concepts. “should” appeals to that which is moral, “pebblesorter::should” appeals to that which is prime. The pebblesorters should not have killed him, but they pebblesorter::should have killed them.
Think of it this way: imagine the murdermax function that scores states/histories of reality based on how many people were murdered. Then people shouldn’t be murdered, but they murdermax::should be murdered. This is not an endorsement of doing what one murdermax::should do. Not at all. Doing the murdermax thing is bad.
You didn’t understand what TheOtherDave said. He was talking about the same usage you are talking about and commenting that it is in contrast to Eliezer’s past usage (and past advocacy of usage in conversations about how he uses should-related words.)
My name is TheOtherDave and I endorse this comment.
Ah, whoops.