California Trilogy: City of Angels - On Camera (Game Over)

[GoofballsAndBaloons]

SL didn't ask the question yet, merely said that he WILL ask the question.

Any list of random numbers is worth any other list of random numbers. I happen to trust that SL does not have knowledge of the number's meaning and therefore cannot manipulate them.

As for who goes last, it's to ensure that if there are signs of something not kosher going on, we have a townie double-checking. We need someone we reasonably trust to be a townie to do this. The last person going in cannot be double-checked, and has the power to double-check others. It's very dangerous in the hands of presumed scum. I can think of several things they could do with this, but I won't elaborate as I don't want to give them ideas.

In fact, it would be best if SL went last. It's a key spot.

[Thok]

"SL didn't ask the question yet, merely said that he WILL ask the question. "

He said he will be asking the question NOW. That implied that he had sent in the PM. (It is possible that SL can delete the PM he has sent before Mr. Grey has a chance to respond, after seeing your question. It's not something I'd stake this decision on.)

"Any list of random numbers is worth any other list of random numbers."

Only if you make a plan before suggesting the numbers, otherwise you can shift the plan to take advantage of the numbers.

[GoofballsAndBaloons]

If you think my plan sucks, why don't you suggest a better one, genius?

[ShadowLurker]

Already asked the question, the set {74, 14, 39, 23, 51} does NOT contain at least 1 very good, meaning we now have an empty baseline for other sets.

[ShadowLurker]

Isn't it more optimal to ask for 4?

Does set {74, 14, 39, 23, 51, 18, 52, 37, 69} contain at least 1


If Yes, we break it into 2 {74, 14, 39, 23, 51, 18, 52} contain?
If yes, {74, 14, 39, 23, 51, 18}, If yes, pick 18, if no pick 52
If No, does {74, 14, 39, 23, 51, 37} contain? If yes, pick 37, If no pick 69
Then we have 4 questions to entirely repeat this process with different people.


If the addition of a set of 4 new numbers STILL gives no very good, then we simply add on the next four. At any point when we discover one, it merely takes two questions to find the exact very good.

[GoofballsAndBaloons]

Asuming no one lies. I'm not sure I understand this plan, can you please show all your calculations.

Is there a plan to catch liars? Is there anything to stop scum from lying? Does it matter with that scheme?

[GoofballsAndBaloons]

Don't forget, we have the scummy players going LAST, which means that no one trustworthy can re-ask their questions. Whether we choose to do this or not, if the threat doesn't exist (because we actually have scum in the last positions), we risk being mislead. No?

I must be missing something as usual. I always do.

[VP Baltar]

Then let's change the order if need be. Put the liar liars in the middle of the order.

[Mighty Orbots]

Dweezil, who had apparently slipped away earlier, reappears with another plate of chicken

I like this plan because it doesn't involve lots of screwy math. The simpler the better I think.

[GoofballsAndBaloons]

Regardless of which plan we end up with, we need a presumed townie going last. I'm disappointed that my preferred choice, SL, cannot make it, but I'm satisfied with MO's pro-town performance enough for him to hold that key last position.

Any objections to MO going last?

[VP Baltar]

MO going last is cool with me. That bumps me up next then, correct? What is the question I need to be asking?

[VP Baltar]

I mean, Dweezil....I was distracted by the Monumentally Overbaked chicken.

[GoofballsAndBaloons]

You guys are still acting like we're in a low-end sports bar.

I have a vague memory that we are supposed to be playing bingo in a church basement or something.

[VP Baltar]

I'M ON A BOAT!

[GoofballsAndBaloons]

Good for you...

This boat is real!
I'm on a MOTHAFUCKIN' BOAT!!!

I effed a mermaid!

Playing bingo with effing Carrie FISH-er!

[Mr. Grey]

ShadowLurker has been appointed Stuntman.

[Thok]

Sigh. Goofballs, we did have a plan that was established. There was no reason not to bring up you counter plan before we started.

Also, rereading your original plan it's clear I've misunderstood it. It actually fails in lots of cases; if in your example 74, 18, 63, 34 were all very goods, then we'd get a yes answer to every question. This would also happen if 14, 52,47, 5 were very good. Those two sets don't have any overlap. (I think I can find a lot of other sets of very good that work.)

It's clear we've got a lot of freedom given SL's answer. Just pick a plan and make it work.

[GoofballsAndBaloons]

Godwin,

Only a scumbag would worry about an excessive amount of VERY GOOD's in any given set. Are wou worried about narrowing it down to the

correct

VERY GOOD???? Like they're not all equal? Jesus H Christ. Not to mention how unlikely that scenario is.

Does this ship have a plank?

Oh yeah, and your "girlfriend" from Folsom says he misses you. The FBI is still looking for you, too. I could make a phone call.

Yeah I'm cranky but I haven't gotten any with Unit 02 since I caught the H1N1 flu.

I think I need a coconut passiona fruit cocktail in a carved out pineapple with dry ice smoke coming out of it, a maraschino cherry, and a paper umbrella.

[Thok]

4 very goods in a set of 17 numbers (which is all that my examples require) isn't excessive. The average amount is 17/5=3.4.

I want a solution to the problem that actually tells me what number to pick once I get the answers. The divide by two method flat out says "This is the number that is a very good" once we ask the six questions. Your method doesn't seem to do that even if we get yes in some of the seven questions, which means the failure rate is more than the 1% you proposed.

Basically, it's the same flaw as I pointed out in ShadowLurker's original proposal.

What we should do is the following:

Take a set of size 2^n., where n<=6. For half of the set, ask whether or not it contains a very good. If it does, we keep that half of the set, if not take the other half of the set. Repeat n times. At the end, we have a very good as long as the original set contained one, and we can use the remaining 6-n questions to test for accuracy.

If we take n=6, we have a guaranteed chance of find a very good in the set of size 64, but we have no test for accuracy.

If we take n=5, we have a slightly smaller chance of finding a very good in a set of size 32, but have one questions for checking accuracy.

[hewitt]

So yeah I'm just pretty much going to wait around for somebody to figure it out and then tell me what to do, sounds good?

[VP Baltar]

So far, I like SL's plan the most. Even if someone lied at some point, I think we would figure it out very quickly.

[Talilan]

Death wishes he weren't so busy at work so he could give due attention to the problem at hand... but what can you do... Humans keep getting ill or old or having accidents and Death must attend to them...

But Death is nearly finished collecting the current crop of souls for the week, and anticipates being able to TAKE A GOOD HARD LOOK AT THE M-F-ING BOAT soon.

[ShadowLurker]

Mr. Godwin, for the initial questions of my plan, there are a lot of opportunities to double check. If for some reason we are getting no luck at all, then later we can switch to only asking about two further numbers, which requires one less question to double check.

On that note, I need to relieve myself in the bathroom.

How much longer do you guys think you will need me here? I hear the lines for the bathroom are very long, bingo night always stimulating the bladders and all.

[VP Baltar]

Kelly, I'm ready to ask my question. Just ask the next four in your given numbers?

[ShadowLurker]

Yep