(no subject)

[lhexa]

A miscellany, in lieu of a real entry:

I (finally) watched The Last Unicorn, and really liked it... unfortunately, it fell short of its potential, probably because of its brevity. It was much deeper (read better) than your average Disney cartoon, though.

On the first day of my Numerical Methods class: "My name is Doctor Hugh. This is Doctor Hu."

A few days ago the front door buzzer started going non-stop. In an effort to disable it, I started disconnecting one wire before noticing that tugging on another shut it off. While I was at work it came on again, and my mother cut the first wire. It was the phone line. I spliced it back together the next day, though.

For once, I'll pass on a meme...

1) Reply and I'll respond with something random about you.
2) I'll tell you what book reminds me of you.
3) I'll pick... er, no I won't.
4) I'll pick a (type of) place I'd most like to go with you.
5) I'll say something that only makes sense to you and me.
6) I'll tell you my first memory of you.
7) I'll tell you what animal you remind me of.
8) I'll ask you something that I've always wondered about you (you don't have to answer).
9) Don't feel obligated to post this to your journal.

And finally, who needs drugs when you have Super Monkey Ball?

Comments

[lhexa.livejournal.com]

Well, the number of equivalence classes of Great Foos would depend on what equivalence relation is being used. Would that be the ability to create and/or view some subset of reality?

I can't pinpoint why a badger seems right.

[http://users.livejournal.com/_dw/]

(Heh, I was going to open this with "Well" too, but saw you had done so :) )

The best way of seeing what relation is to be used would be to consider it as regards to the creatures holding the idea (or the surrounding ideas, if the creatures are an invention; for the point, it only changes the semantics).

The Great X is that which embodies the subjective X-ness maximally without degrading. In "reality", the process of trying to get closer to X-ness also brings to a greater clarity what exactly X *is*.

X-discovery as a process cannot be embodied in an equivalence relation, so let's only consider the final results themselves. Then at first, it would seem you can have Great objects for any property that can be embodied by an object, and that limit will probably be property-bound instead of object-bound (or rather, that properties that cannot be directly assigned an object will be easier to discover cannot be such than that of objects that contain properties you don't know of yet).

Thus, how many Great Xs exist depends on the nature of self, and of the universe. Is it possible to encode the relation so that it reflects and makes formal that "linkedness"?

(Searchers are kind of special, perhaps because it's what I've been focusing on, trying to think what would "contain" Searching more than anything else. Technically, anything that can compress a lookup operation into feasible terms of time and space will be a good searcher, and therefore a hypercomputer would be a better Searcher than something that only solves PSPACE-like What-If/strategy problems (wrt logic or the unknown of the universe), and therefore the term "Great" Searcher might be a misnomer since I don't know if hypercomputers are possible. But presumably a real Great Searcher could bootstrap any other Great properties by well, searching for them, the importance of the journey notwithstanding.)

[lhexa.livejournal.com]

I think you just argued fairly strongly against there being a finite number of equivance relations, if the relation is "embodies the same qualities maximally". For one thing, there are different combinations of rankable qualities, so that a Great Searcher could be divided into a Great Information Gatherer and a Great Information Analyzer (or Great Telescope and Great Computer if you want to be prosaic). For another thing, there might be infinitely many properties that can be ranked on merit in different ways. If there were a finite number of ways of ranking merit regarding a property, one could base an equivalence relation off that, but it might not be possible: you have such widely different things, according to your idea, as a Great Digger and a Great Sorter.

Do you know what an equivalence class is? 'Cause I'd have thought you'd bring up the trivial equivalence relation.

[http://users.livejournal.com/_dw/]

Well, if the greatness reflects the nature of the universe (in a sort of "objective Platonism that colours subjective interpretation"), then there might be a finite number of Great properties, at least to the extent that properties that are combinations of simpler ones will be discovered as such. In that case, the relation would be something like "is in the same position in the universe's hierarchy of properties".

I don't know if properties can merge, and if so, their "aspects" would be considered a "great-of-sorts aspect". For instance, a Searcher is both a computer and a telescope, or can work as both. As a computer, it's the ultimate constraint programming machine. As a telescope, its "find the right dimension" property means you can see anything that exists (limited to what you could in theory see?).

Do you know what an equivalence class is? 'Cause I'd have thought you'd bring up the trivial equivalence relation.
It's a bit unclear to me, but I think it is a class of elements that are equivalent to each other under the relation.

[lhexa.livejournal.com]

Oops.

Yeah, a relation concerning position in a hierarchy could work, but to get to it we'd have to construct the hierarchy.

No, an equivalence class is just a class whose members are all related to each other under the equivalence relation. If they're related they're automatically "equivalent". The trivial relation is just "...is the same as itself, as is...". It gives you one equivalence class that holds everything.

Nontrivial equivalence relations involve things like unit conversion and set size.