Public: Election: Random
Increased Entropy Government   (+6)  [vote for, against]
Use controlled randomness to increase rate of governmental evolution

Large governments change very slowly over time. Generally, it takes a catastrophic event - such as revolution - to change a government completely. A government is a complex adaptive system (get ready for a paper-based reference (sorry) "The Quark and the Jaguar" by Murray Gell-Mann). Complex adaptive systems are systems that adapt in response to their environments. A child learning to read is an example, as well as the evolution of a butterfly. The best adaptive systems include some degree of randomness in making decisions. This allows the system to try out different solutions to a problem that may not be immediately logical but might lead to a better solution than a solution based in logic would have led them to. If you're a tree-climbing animal being hunted by others, the best next step might not be the status-quo "evolve to climb better than my competitors", but might be in a completely different direction - "randomly evolve a skin flap (hi [skinflaps]) to fly between trees".

The state and federal governments in the US, and I suspect elsewhere, are painfully rule-based and stagnant. Right now if the Democrats want to pass a law they have to somehow bribe Republicans to go along with them by trading other political favors. Laws aren't voted on by their social merits but instead by their political merits. The same is true in elections. The best candidate doesn't always win, it's usually the person with the best campaign, which is often the candidate with the most money.

I propose the following system. Instead of majority rule, everyone places their vote in a big drum. The drum is turned a few times, then a ballot is pulled from it. The vote cast on that one ballot wins. This keeps our representitive democracy - if 29% vote for Bush, he still has a 29% chance of winning - but allows for randomness. Third party members will win every now and then. Even every once in a while some real extremist will win. That's ok - it's part of the system. If their policies are looney, we can probably fix them in the next term. On the other hand, what if this fringe politician understands something that nobody else does? Radical change could happen.

Granularity of this concept could be controlled. If we want large changes across the board, implement this at the election level. If we want smaller changes, then only implement this at the legislative level.

I believe the core concept isn't new. Didn't ancient Greece have a similar system of white and black balls that are placed in a box, and one ball drawn? I don't know - wish I had payed more attention in history. Anyway, to my knowlege this system hasn't been implemented on any large modern government in the way I've described. If I'm wrong, I'm sure someone will mention it.

<AddedLater date="Mar 14 2003">I just thought up another way of varying granularity. Several people have expressed concern over some crackpot being elected. In fact, I could vote for myself - hoping I'd win the presidential lottery. If only one card was entered for every, say, 50,000 votes then you'd at least eliminate the severely extreme candidates. This number of votes can be varied to set the granularity. 1 card / 1 vote = my original system (high entropy). 1 card / 50 million = current US system (low entropy). Wait, that would have made Gore the winner.</AddedLater>
-- Worldgineer, Mar 11 2003

The sad tale of Dr. Thomas Butler http://www.cbsnews....es/main578660.shtml
A salutory lesson in the misuse of power. [DrBob, Oct 04 2004, last modified Mar 16 2006]

//The best candidate doesn't always win//

How do you know?
-- beauxeault, Mar 11 2003


Notice the current American president. Case and point.
-- Worldgineer, Mar 11 2003


//You made no case or point.//

Oh, stop it. Can we please assume sometimes someone gets elected that isn't perfect for the job? Or even if they are logicly the perfect person for the job in the end there could be someone who has better ideas? I'll even delete my bush posting if you'd like, just let's get back to evaluating the acutal merits of this idea.
-- Worldgineer, Mar 11 2003


"Can we please assume sometimes someone gets elected that isn't perfect for the job?"

Weeelll . . . okay.
-- bristolz, Mar 11 2003


I'd say it happens 100% of the time. That doesn't mean the "best" candidate didn't win. The most objective way to measure a candidate's "best"-ness (that is, among those who are competing) is to hold an election.

On the merits of the idea, I think it poses some intriguing possibilities/opportunities, particularly with regard to achieving revolutionary change when the time is right. But I don't think I'd be willing to pay the price when the time or the idea isn't right. I'd also have to think that your proposal would, almost by definition, elect the "best" candidate less frequently than the current system.

I understand how the randomness helps butterflies to evolve, but that doesn't mean the process is not sometimes very painful to butterflies, nor that it's one butterflies would impose upon themselves, if they had the choice.
-- beauxeault, Mar 11 2003


[dag] //If an extremist did win, at least in the U.S, the damage could be limited. Other countries, whose leaders have more solitary discretion might not fare so well. But anyways, the current (U.S) system, altough not perfect, works. So why change it?//

That's where granularity control comes in. In a more total-control country you speak of it may work better in the legislature for voting on specific bills. This will virtually get rid of the type of political favor trading that goes on, as the result isn't controled by getting that one last vote. //The current (U.S) system... works// I would disagree. I guess it depends on what you mean by "works". I propose the current federal system is stagnant, burocratic, and rife with legal bribery. Sure, roads get built, but only if the senator from your state has politically manouvered into the right committees.

[beau] //The most objective way to measure a candidate's "best"-ness is to hold an election.// Exactly - you're going by what can be measured, observed. The current system is very calculating and rigid, which I would argue is arbitrarily filtering in nature. It allows for no variation, no creativity, no surprise. Any system this rigid misses anything out of its narrow view. A higher entropy system would still keep democracy at the forefront - your candidate has the correct odds of winning. It just widens the possible outcome.

[raven] //I'd suggest trying this in a small town first, and only if the towns nearby agree to take in refugees if the person elected turns out to be a total loon.//

//You don't want to give the randomly-elected too much power. President would be a bad choice. I wouldn't be against picking a senator or two this way (since the randomly-elected ones will still be outnumbered by the traditionally-elected ones). //

//Actually, this is reminiscent of "Mr. Smith Goes to Washington". //

Maybe you're right. I'm feeling a lot of resistance to the elected president concept, which I'm fine with. A society that enacts this at all levels may evolve very quickly, but may also crash just as quickly. We don't have to jump to the most granular level first, maybe start small like with bills in the legislature.
-- Worldgineer, Mar 11 2003


[raven]//<totally and incredibly off topic, long, and opinionated> Actually, [World], in my opinion, the biggest problem our government has is riders (the act of attaching an unrelated appendix to a bill before it is voted on). You get a good bill with a bad rider attached, and those who represent us have the unenviable choice of either voting for the bill and taking the bad rider along with it, or voting 'no', even though they really wanted that bill to pass. There was a wiley senator once who successfully blocked a bill by attaching a rider to it that would have made it illegal for trains to move. This situation has been partially corrected by giving the president the power of 'line-item veto' -- the ability to sign a bill but waive its riders -- but until congress has the same power, or riders are eliminated altogether, things will continue to be a mess. (Yes, I know that this is not the ONLY problem with the system. But in my opinion it's the biggest, and we should fix this one first, before moving on and fixing everything else.) </>//

<responding to off topic comment, which isn't off-topic after all>Ah, but riders have to be voted on in committee first anyway. This is just one aspect of the political favor trading that goes on - you vote for my trailer, I'll vote for yours. The solution? High entropy voting in committee votes. Less incentive to trade political favors if you're not relying on that last vote, but instead a bit of random chance. I'm not saying it will completely solve the problem, but should help</rottcwiotaa>
-- Worldgineer, Mar 12 2003


Entropy....US Government... *consults the Department of Redundancy Department...
-- RayfordSteele, Mar 14 2003


Are we not currently suffering problems due to a crackpot being in charge of major country? However in the case of this crackpot he was selected rather than elected so this does not necessarily cast a shadow on democracy.
-- Aristotle, Mar 14 2003


[raven]// Re: [<AddedLater date="Mar 14 2003"> stuff]

Actually, after writing [my comments of March 11], I realized that we already have a system in place to keep crackpots out. You need a certain level of support (measured by obtaining a certain number of signatures on a nomination petition) to get onto the ballot at all. Of course, the occasional crackpot still manages to slip in... //

...and get elected president.
-- Worldgineer, Apr 21 2003


Currently, the normal strategy used by politicians is to do whatever it takes to be considered "best" by about 55% of the populace, and let the other 45% be damned. After all, a politician is no better off with 99% of the vote than with 55%. Adding a certain amount of "dither" would encourage politicians to work toward more broad-based support.

Point to consider: were it not for the Electoral College, a politician who could pull 90% poll results in Chicago, New York, L.A., and a few other cities could win the presidency, even if everyone else in the country voted 100% against them. Under such an environment, politicians wouldn't even bother campaigning in 90% of the country. That would decidedly not be good for the republic envisioned by the forefathers.
-- supercat, Apr 21 2003


Wow. A great idea. The reference to the ancient Greeks makes me think we might have ended up with the wrong method of democracy!

Think how much cheaper it would be to run an election 'lottery' style. You don't count the votes, just collect them for the draw. The result would be MUCH more television-worthy, too!

To avoid crackpot leaders, I would want at least 100-500 votes drawn from the lottery.

The lottery style would also be an added incentive to voting. That is, think about the relative psychology of 'it might be my lottery ticket' over 'I am just a drop in the voting ocean'.
-- not_only_but_also, Apr 21 2003


[noba] I like the TV aspect you have in mind. Maybe on your vote you can check a box saying you give permission to use your name if your vote is one selected.
-- Worldgineer, Apr 22 2003


Fine, as long as the elected leader can only rule for 29% of the time that a normal leader gets to.
-- phundug, May 21 2003


Why?
-- Worldgineer, May 22 2003


Maybe try this out for congressional elections first. Pulling a couple hundred votes out of a couple hundred barrels in different states would add another statistical layer to catch the outliers/anomolies/wierdos. You would increase the chances of getting one, but they effects of getting one would be minimized. Also, you would have to add some sort of mandatory training period for all elected due to general ignorance (myself included) in the mechanics of how congress operates. Or doesn't.
-- schwantz34, May 22 2003


[sch] I think the education of congress is an entirely seperate issue.

[Ob] There is no protection against that currently (see US present administration). The proposed system should aid in the evolution of government, but I do not claim it is a magical system that can keep idiots out of office.
-- Worldgineer, Apr 12 2004


[Wordlgineer] while I was hoping to leave political shots aside, as I've stated earlier, this is a Newtonian reaction, so...

I'm finding it comical that in idea about the evolution of government, you're actually taking shots at an administration that is considered radical by its opponents and in fact attempting to implement radically new policies such as preemption.

And it's an administration that has presided over more changes to our government (regardless of the merit you assign them) then any other since WWII.

As to your idea, the party system introduces rigidity as the price of "keeping radicals out".

Rather than using an absurdist (in my opinion) approach of randomly choosing amongst candidates, it would be better to advocate more "independent of party machinery" candidacies -- which the Internet, witness Howard Dean's early rise, can indeed fuel.

If we for instance could directly vote for President among the group of candidates in the primaries in 2000, it's likely that McCain, not Gore, would be President today.
-- theircompetitor, Apr 12 2004


I like this idea. There's too much inertia and too little choice in the US current system.
-- Enjoyer, Apr 12 2004


[tc]//taking shots at an administration...// I'm just pointing out that an even in a low-entropy system we are not safe from forces that are counter-productive. I'd argue we are far less safe from such forces in the long run.

//advocate more "independent of party machinery" candidacies// I find your approach to be short-sighted. I think you'll find that any method such as directly reaching the electorate through the internet will quickly be swallowed by the political machine. The current system is self-adapting with the goal of keeping those currently in power safely in power in the future. Rules built into our constitution help fight against this issue, but are more of a band-aid than a cure.

//it's likely that McCain, not Gore, would be President today// Ha, thanks for the admission. However, your point has little to do with my idea (other than pointing out one of the thousands of rules that make our government rigid and stagnant).
-- Worldgineer, Apr 12 2004


//Ha, thanks for the admission// just checking to see if you're awake :)

Political machinery evolves quicker than our ability to cut off the vines -- witness the Democrats use of soft money in this election, forgoing of matching funds by both Dean & Kerry, in spite of McCain/Feingold.

I think the reality is that in most cases independent candidates just don't have much to add, and worthwhile non-independents, like McCain, have too much to lose to jump on that wagon.

But I do believe the Internet can change that, simply because fund raising is easier, witness Dean.
-- theircompetitor, Apr 12 2004


<metaphor blender>//Political machinery evolves quicker than our ability to cut off the vines// Exactly my point, and how we've come to our current situation. Instead of trying to cut off vines (changing a piece or two of our political process), let's remove fuel for the machinery (the incentive for deal making).</metaphor blender>
-- Worldgineer, Apr 12 2004


hmm... I think I was going after the same goal in my "draft first term representatives" idea.

I don't know if people would be comfortable doing that with as high a profile job as the presidency -- witness the outrage over the quirk in the last election -- in a sense, the butterflies provided the extra randomness.

Also, I've always thought that changing how Congress works is the key -- why do you think that changing the presidency would make any difference? Or is your rule to be applied across the board in all elections?
-- theircompetitor, Apr 12 2004


I never mentioned the presidency, except in an example. This idea is aimed generally at elections and votes.

I would personally advocate for my original system to be implemented for ballots in the general senate as well as in committees. I would implement my amended version for elections, with local elections having low entropy and national elections having a medium level of entropy.

//witness the outrage over the quirk in the last election// The problem was bias, not randomness.<aside>clever butterfly comment, [tc]</aside>
-- Worldgineer, Apr 12 2004


The system was supposed to incorporate this aspect in its inception. Since then, of course, things have gotten seriously rotten.

One method of simulating this is "apportioned voting." Instead of limiting votes to one candidate per voter/contest, each voter could vote for any and/or all candidates in a contest. 4 people running and you like what 3 of them have to say? Vote for those 3 and ignore the one who you don't like. Feel the need to vote only because of patriotic or other obligation? Vote for everybody - you're self-cancelling yet still performing your civic duty.

At the end of the day, the candidate with the most votes still wins, and a much greater accounting of attractive options is also determined.
-- corquando, Apr 14 2004


Though I like the concept of apportioned voting, it has little to do with my idea.
-- Worldgineer, Apr 14 2004


//The problem was bias, not randomness//

If you're referring to the Court's decision, that's why the court has an odd # of judges on it. The constitution does not require unanimous decisions or even supermajority decisions, and that's a good thing for the "evolution of government".
-- theircompetitor, Apr 14 2004


I was talking about the butterfly ballots, hanging chads, legal gaming by both sides, political and familial connections, errors in not allowing qualified voters to vote, and many other real or perceived sources of bias. I don't think anyone would mind these issues if they were random and effected all sides equally.
-- Worldgineer, Apr 14 2004


I am in favor of anything that controls granularity. Goddamn granules!
-- bungston, Apr 14 2004


[BEA], very interesting article in the link. I think I'll start using their definition when someone asks me about entropy: "entropy is a measure of the uncertainty in which microstate will be observed in the next measurement" (although "microstate" should probably be "a set of microstates"). Rolls off the tongue. Anyway, thanks for clearing up any confusion some might have. Again, I highly recommend reading "The Quark and the Jaguar".

[bung] I think you mean "Goddamn grain!" You perhaps have stumbled upon my aspirations for an Atkins-like low fiber government.
-- Worldgineer, Apr 15 2004


Let us say the job is security guard. Do you hire someone hoping that they will do the right thing and be at their post? Or do you have procedures in place to disipline or fire them for poor work performance? It is just commonsense. I think you may need to re-examine the problem. As it stands go fish.
-- Around TUIT, Aug 13 2004


I don't think you understand this idea at all. Or I don't understand your comment. The same rules would be in place for pulling someone out of office.
-- Worldgineer, Aug 13 2004


This idea goes too far in trying to keep the wrong person out that it makes it harder for the right person to get in. And a government isn't a living thing, we are. Nature rules whether we want it to or not. Governments, organizations, and the like are an attempt to conteract it. One of the advantages in this is they are not alive. They are not subject to the same things we are. Papers like the one you referenced are our way of understanding social structure comparing it to what we know best. Repetetive changes in state observed in our enviroment. Nature.
-- Around TUIT, Aug 13 2004


//This idea goes too far in trying to keep the wrong person out that it makes it harder for the right person to get in.// There you go judging who the "right" person is. With the current rigid structure, I'd say it's near impossible for all but a certain type to get in. I sure hope you're correct in calling them the "right" ones.

Re: (points about government is good because it's not nature): Government is currently what is called a complex adaptive system, just like the learning abilities of an animal's brain and like evolution. You generalize and call this nature. Fine. But then you tell us that government is created to counteract nature. Certainly you are mixing up definitions, unless you are saying that the point of government is to keep from reacting to changes. I see the reason for government is to use the collective powers of people to change our collective actions in a desirable way. Optimizing this process can only enhance this goal.
-- Worldgineer, Aug 13 2004


"right" as in the "right" security guard. One who doesn't the post, fall asleep, etc. Your idea doesn't improve enforcement of expectations of those who are being represented.

On the second part of the comment: comparing two things doesn't make them the same thing. All of the things that are differences are ignored to make a point about the similarites. What I am saying is people compare things in nature (butterflies,brains,whatever) to things not found in nature (types of government, tech, etc) because it is easier to understand. It also helps justify your idea because the thing in nature actually exists. One form of method is called personification. Also I never said one is better than the other. Let me clearly state Nature is clearly better because it is existence (EVERYTHING) and I love it. But in the case of government with regulated farming, resources,fighting disease is not a policy of letting nature take care of it. Which I see we both agree with. Funny though how you would trust nature with your leadership.
-- Around TUIT, Aug 13 2004


//improve enforcement of expectations// What's this about? If you mean that this idea doesn't ensure that people do their jobs, well you're right - it doesn't claim to. The current system has no greater incentive or enforcement that politicians do their jobs than this.

//comparing two things doesn't make them the same thing// Fine. But your arguement isn't making sense to me. I'm using nature as an example to help people understand what a complex adaptive system is, not to necessarily compare it to government.

//trust nature with your leadership// I still don't know what you're talking about. Are you saying nature = randomness? I'd much rather have randomness in politics than bias.
-- Worldgineer, Aug 13 2004


//What's this about? If you mean that this idea doesn't ensure that people do their jobs, well you're right - it doesn't claim to. The current system has no greater incentive or enforcement that politicians do their jobs than this.//

I was responding to this statement //There you go judging who the "right" person is. With the current rigid structure, I'd say it's near impossible for all but a certain type to get in. I sure hope you're correct in calling them the "right" ones.//. And if you don't feel that there isn't a problem with our concepts of leadership why are you proposing an idea to mess with it? Where's the problem or lack of anything you are trying to address? It doesn't matter what the people in charge call themselves if there are safeguards in place to make them respond to the people they represent. If you owned a business would you spin a wheel to see who's your new cashier? Under your idea the KKK could of been running the United States (David Duke)! Screw waiting that out! What about the pro-slavery corporate backed party? I'm sure they could get a lot of votes in the barrel with some slick advertising. That's what I mean.

//I'm using nature as an example to help people understand what a complex adaptive system is, not to necessarily compare it to government.//

you're using nature to help people understand what a complex adaptive system is and then comparing the complex adaptive system to the government are you not? nature = complex adative system = government Isn't that the base theory of your idea?

//Are you saying nature = randomness? //

We often call unknown things randomness because we don't understand them or see a pattern but observe a change in state. Nature is. Randomness is a just label we use to understand part of it.
-- Around TUIT, Aug 13 2004


[Around] Sorry I missed your anno - I've responded below.

(everyone else) Sorry about all this chatter.

//if you don't feel that there isn't a problem with our concepts of leadership// I think there's a problem with our structure of govenment.

//If you owned a business would you spin a wheel to see who's your new cashier?// That's a poor analogy. Perhaps if there were a million store owners, and each one put their vote in, and for each hundred votes a name is entered in a drum, then pulled the name from a drum. Then yes, I would.

//Under your idea the KKK could of been running the United States// At the highest entropy level, yes. I suggest a much lower amount of entropy so that things don't change as much as that.

//nature = complex adative system = government// No. (some aspects of) government and (some aspects of) nature are complex adaptive systems.

//we often call unknown things randomness because we don't understand them...// I still don't understand what you're getting at here. The distinction between the two is unimportant when talking about putting votes in a big drum.
-- Worldgineer, Aug 30 2004


I am a disillusioned anarchist, but really, seeing what happens with the actual system, this can not possibly be worse. Let's try it. I like it. Or perhaps we could croissant and fishbone our politicians. What do you say?. Croissant for World, fishbone for Bush (I know, I know, It's not my pain in the ass, or perhaps it is).
-- finflazo, Aug 30 2004


"Didn't ancient Greece have a similar system of white and black balls that are placed in a box, and one ball drawn?"

[worldgineer] I think you are referring to the Hellenic tradition of ostracizing citizens. All citizens had to cast their vote, but they introduced one "white" vote in the drum, in order to incorporate the different notions of "fate", "luck" and "chance" into the process. As you know, the ancient Greeks loved the concepts of luck and fate. The person who faced being ostracized could still be saved by the joker, so to speak.

Moreover, the Greeks already thought that their politicians embodied something of the "randomness" of life. The Greek ideal of "Kairos" which denotes the ability of a person to seize an opportunity at the right moment, and which involves a complex play of luck, blind fate and agility, was hailed as being the most important quality of politicians.

I think this idea still holds true. You are faced with a stagnant system of rules, but the game which developes within this system is full of indeterminacy, coincidences, bizarre shifts in alliances and irrationalities.

I don't see how your idea would change this. What's more, I think your idea is an accurate description of how politics work in modern democracies anyways.

But it's still a nice description.
-- django, Aug 30 2004


The difference is bias. You're seeing a game played out inside of a small box. The bias in the system keeps the game inside this box, when real answers may lay outside. Removing the incentive for backroom deals and aliances, as well as allowing for an unpopular law/candidate to win every now and then widens or removes these walls and makes more outcomes possible while keeping the popular outcomes likely.
-- Worldgineer, Aug 30 2004


Wow, [world] - I thought it was a [vernon] idea until I reached the end! [+] anyway for suggesting we try to improve our election systems. The idea makes more sense the more you increase granularity, although it's kind of like taking the Single Member District system to the extreme. I notice there's a lot of this sort of idea about - wonder if Bush/Blair may have anything to do with this...
-- wagster, Aug 30 2004


//You're seeing a game played out inside of a small box.//

True, but aren't you simply creating a new box (the rules of which are: "let's all agree that we use controlled randomnization to increase governmental evolution"?)

Your idea would only be complete if, In The Beginning, there is a "rule" which says that at random intervals, the entire system (of randomnization) itself gets changed in random ways - and this Founding Rule itself must be of such a nature, that it randomly whipes out its own history (which would always point to an origin of consensus - the opposite of randomness), so that nobody will ever know that there was a "rule" at the beginning of your randomnization process. You need some kind of "autopoietic" mechanism.

And I'm not sure where you will find it.

[This is all over my head.]
-- django, Aug 30 2004


But that is the case! You've described the history of government. It doesn't perfectly conform to your model, with each iteration keeping parts of the last set, but a good adaptive system should keep pieces of the last iteration. It was a long slow and painful process, and we've come a long way. I'm not talking about starting over, just adding a mechanism for controlling our rate of evolution.

//[This is all over my head.]// Clearly it is not. Let me know if there's anything you're not getting though.
-- Worldgineer, Aug 30 2004


The mistake in the idea, World, is in correlating getting elected with having power. In your random system any fringe candidate who managed to stumble into office would pretty soon fall foul of the Senate, House of Representatives, House of Commons or whatever and almost everything that the new president tried to do would be blocked and, eventually, they'd be organised right out of the loop of government altogether if not actually kicked out of office.

This was demonstrated nicely by the British Conservative party a little while ago where Iain Duncan-Smith, the first Tory leader ever to be elected by the party members was evicted from his job by the simple expedient of the party MP's refusing to work with him and then agreeing between themselves who his replacement would be.
-- DrBob, Aug 31 2004


True, though you'd need more than just a majority to be against them. Almost all of the horse-trading power is gone if votes are randomized. If just a few Senators/Representatives/etc. voted on their bill they'd have a chance.

Actually, thanks for showing that even at the higher entropy levels we still couldn't have a madman completely controlling things due to the internal checks-and-balances systems built into govenments.
-- Worldgineer, Aug 31 2004


I'm not so sure about that last paragraph. If a madman, or even someone slightly erratic, can recruit enough mad or ambitious friends to his/her cause then they can still run things. I'm not a great believer in systems as a means of stopping people doing what they want to do. Systems can be broken no matter how robust they are if enough determined people hold critical positions in government. And, as history demonstrates, those systems can then be turned against those that they are meant to help or protect; I refer you to the current, rather salutory case in the USA involving Dr. Thomas Butler (linky).

The only real control on politicians is exercised by the population and it's willingness or otherwise to take organised action against a government that they don't like.
-- DrBob, Aug 31 2004


Great yet terrible link, [Bob]. Yes, I agree we have a whole team of madmen in office right now. I don't guarantee that my system will change that. In fact, if people keep voting for madmen then they'll likely keep being elected into office. Unless a non-madman gets lucky and his/her card is drawn.
-- Worldgineer, Aug 31 2004



random, halfbakery