Time Heals All Wounds.. And Then Kills the Patient
<Previous Next>
Evening
Evening
Sun May 2 16:19:59 2004
Undeserved Gifts

Notion::(In_Quantum_Physics::Randomness) Assertion::(Previous::Allows_Free_Will) Counterargument::(Randomness::Unlike_Choice) Counterargument_Supplement::(Randomness::Unlike_Rationality) Counterargument_Counterargument::(Nature_Of_Randomness::Not_Without_Regularity) Counterargument_Counterargument_Supplement:: (Nature_Of_Randomness:: Study_Of_Data_With_Unknown_Regularities_and_Inputs)

Yes, what I'm musing over, again, is, if we were to grant that randomness and statistics are not concerned with things truly random, but instead with unknown causes of variation (A difference that might be seen as akin to how geometry studies abstract things, like circles, that have no existence in the world but are nontheless practical to study), does this weaken this (3)?

Argument map: 1) Free will is nonsense because the universe is, on a large scale, deterministic, and quantum effects may weaken this on the small scale, but on the whole produce sufficiently deterministic results on a large scale (that is, in aggregate) that the system is deterministic at the level that the brain operates 2) The brain may be fine-grained enough that quantum effects directly play a role in its function, therefore the determinism argument is not decisive for free will, and reasonable people may still believe in it, imagining the soul is what fits into those quantum spaces 3) Quantum effects are random, and randomness is very different from notions of free will or the soul

Of course, there are a lot of other interesting possible objections to #3, but depending on one's understanding of the philosophy of statistics, this is a pretty strong route to take. Not being a believer in free will, souls, or consciousness (at least, most of the content -- you might be able to define it in a way I wouldn't deny it) myself, this argument poses an interesting challenge to me. There really are a lot of interesting assumptions and philosophical assertions behind some bits of science that it's fun to take a look at. Unlike, say, my car, when I peek under the hood, it leaves me thinking.

And now, an analysis of this. In case the news story goes away by the time you read it, it's a 2-sided picture, on the left, a picture of Clinton, on the right, of BushJr. The left caption says "I did not have sex with that woman", the right caption says "Saddam Hussein had the materials to produce as much as 500 tons of sarin, mustard, and VX nerve agent.". The bottom caption says "America impeached the guy on the left for lying.". Is this fair? Things that appear pretty uncontroversial, at least to me: 1) Clinton was, if not lying, mispresenting himself in that statement. 2) If BushJr was lying about that and used it as a pretense to war, he belongs in prison. It's much more serious than lying about personal behavior. 3) BushJr's statement was incorrect

What's the problem? We can imagine a number of situations, ranging from the most horrific lie to mistaken belief based on good evidence, that might've been. It's really hard to tell after the fact, given the information we had. Let's present a few scenarios...

1) BushJr gets intelligence reports that mostly concur that Hussein can and was planning to produce such weapons, and used them as a primary reason to invade Iraq 2) BushJr wants to invade Iraq for other reasons, and uses normal speculation and unreliable suggestions to form a justification fo invading Iraq, with the belief that there probably were no such weapons 3) BushJr was eager to invade Iraq, but needed a solid reason, and jumped on slightly convincing evidence as a way to justify his invasion 4) BushJr has a hunch Saddam is up to no good, and uses slightly convincing evidence to fuel a hunch, which he believes will be borne out, that Saddam is producing such weapons ...

Yes, there are a lot of possibilities. So, dear reader, I ask you, how would you judge BushJr in each of these cases? In the case of #2, are there any reasons you could imagine that you would be comfortable with him wanting to invade Iraq under false pretenses? Oil? Concerns for the Kurds? Ending his general abuses?

Speaking of political things that may disgust, Macedonia admitted to a real shocker recently -- their security forces lured some Pakistani immigrants into one of their cities, on promises to ferry them further into Europe, and then were surrounded and shot by said security forces. The Macedonian government did this to claim they were terrorists, and were apparently (!!) aiming to impress the international community with their willingness to fight terrorism. Now the involved government officials are being charged with murder. According to an Interior Ministry spokeswoman, "It was a monstrous fabrication to get the attention of the international community". Word back to Macedonia: We're all really impressed, let me tell you. I guess the pro-America folk are happy not to be the only people with egg on their face. The Muslim world is probably not amused.

Ahh, the undeservedness... Another person with my name wrote a peacenik article, and I'm getting some sympathetic comments from random people who apparently plugged "Pat Gunn" into google and found me (yes, google loves me). While I appreciate her pacifism, I don't deserve the praise that belongs to her. On another note, but in the same key, I got a term paper back from my Cognitive Neuropsych class, and I got a B on it. I'm ok with the grade, but I really don't think the paper was very good -- although I had a kickass title (Cognitive Neuropsychology's Silver Hammer -- was a paper analyzing an attack on Cog Neuropsych as being like hitting a car with a hammer, with specific analyses), my paper's main point was muddled, and my examples were pretty generic. The paper didn't really feel like it *worked*. Oh well.

Oh, you may be amused by the story of this Atkins couple.

I briefly was amused at really big buildings.

If you're using Redhat 9, it's time to get on FedoraLegacy.

I recently read about OptInRealBig's lawsuit against SpamCop. It's not the first time spammers have sued spam blacklisters.. In reading the specific list of charges, and asking Debb about them, I find myself disturbed that some of them are actually real torts. In particular, apparently these are all mostly the same thing: Tortuous Interference with Contract Interference with Contractual Relationship Intentional Interference with Prospective Economic Advantage Negligent Interference with Prospective Economic Advantage

What they mean, apparently, is that a third party, spamcop, is getting involved with and interfering with their contract with their ISP (as well as other ISPs), and preventing them from entering new, similar contracts. No doubt there are a bunch of rules about when this should be actionable in court, but I can't imagine these rules being actionable ever being a good thing. They seem to me to be rules designed largely for the benefit of big business, and they also seem to me to not obey a commonsense and good ideal of contracts -- if you're not party to a contract, you don't have to care about it because it doesn't bind you in any way. A lot of laws and torts, I believe, that are on their face problematic have all sorts of carefully-crafted exceptions and things that need to be met (or not be met) in order for them to apply/be actionable. In this case, I have a hard time imagining any good, nonplutocratic reason for these to ever apply/be actionable. In any case, best of luck to spamcop, and if they fail, we'll need to collectively build a distributed, hard-to-sue-or-shutdown system to squash spam companies.

Oh, Google is starting its IPO. There's an interesting thing to be learned about how stock in corporations works -- apparently companies have a lot of lattitude in how they can set these things up. I wasn't aware of this, and it's a fun factoid. I already knew about preferred and regular stock, and I guess this is a slight variant.

And now, a story. When I was young, I was visiting a friend somewhere, and was playing the old NES8 game Metroid with them. Later, when it was time to Ahh, actually, it was a relative, I think. Anyhow, when it was time for me to head back to Brecksville, I saved the game, and grabbed a pen to write down the password.. and they asked why I was bothering, because they thought Metroid had a battery in it. I told them, no, Metroid didn't have a battery -- Nintendo made sure that game boxen were marked for games that used batteries, and the box for Metroid lacked such a thing. They said that it must have a battery -- how else could all the information in the game be stored in the small list of numbers in the password? I said that I was pretty sure the password I was writing down would work on my copy of Metroid, but couldn't explain how the password, which they thought was a security thing, was linked to the 'saved' game, although I pointed out that that the "JUSTIN BAILEY ------ ------" password always led to the same state. Indeed, now I understand, having done a lot of programming since then, how the password could indeed store all the data needed in a metroid game, with plenty of room to spare. Let's work through it, going on my bare memory. There are 24 spots in the password, and I believe that each spot can hold any letter, one of 10 digits, and at least 4 punctuation characters. That's at least 38 distinct values for each position, and 38*24 is 912 bits of information, or about 114 bytes. We know that the vast majority of the possible codes one might guess don't work, so let's imagine 2 bytes of those are used as a validating checksum, leaving us 112 bytes to work with. What are most of the things in a saved game in metroid? What does 112 bytes need to remember? A few things... Which Energy Tanks Have been found (8 binary values, or 2 bytes) Whether the 'long beam' has been found (true or false, 1 bit) Whether the 'varia' has been found (true or false) Whether the 'jump boots' has been found (true or false) Whether the 'morph ball' has been found (true or false) Whether the 'morph bombs' has been found (true or false) Which weapon is equipped (Normal, Ice, or Wave, 3 values, probably stored in 4 bits) Which Missles have been found (8 binary values, or 2 bytes) Whether Kraid Has been killed (2 values, or 1 bit) Whether Ridley Has been killed (2 values, or 1 bit) Whether Samus's 2nd form is unlocked (2 values, or 1 bit) Which doors have been opened (Maybe 24 binary values, or 3 bytes) Which elevator the save is at (maybe 8 values, or 1 byte)

I may have forgotten a few things, but let's add that up. I get 68 bits, or about 8 bytes. By that calculation, there's a lot of room to spare, as we're using only 1/14th of the total space available with this estimation. In other words, despite the large number of possible states, it all easily is encoded, with room to spare, in the space available in a password in that game. In fact, it would all fit in merely 4 characters and still leave us room for a lot of extra state (if I'm not forgetting anything, it'll fit exactly into 3).

And finally, a last proof-of-concept for the real doubters.. If you really doubt that information can be so encoded, consider the following: Imagine if you were asked to simply enter all the above information in dialogue after dialogue, when you're restoring. Would you agree that that would faithfully restore all the data from your game? If not, what difference could you see while restoring? If anything, we could encode that by adding a few more bits of representation. If so, we could easily could inefficiently encode all 13 of those variables, giving each of them one of the 16 positions, and assigning each value a particular character.

Anyhow, this explains very easily how this purity test can encode, in it's results all those values in just 6 numbers. (note that those arn't my results, it's just 111111).

Finally, you *will* enjoy these: Beef A Lo