Super Goo

Scott Hardie | April 9, 2008
I GOT IT!!

FINALLY.

I've been kicking this around in my head for some time: How to make a scoring system that isn't random and isn't time-based, but is still able to separate a winner from within the class of hardcore players who never miss, and ideally that would involve some kind of grid or graphical representation. I think I finally have it.

All 50 goos are created in advance, 49 regular goos and 1 "super goo" (actual term tbd). You begin with a 7x7 grid containing all of the regular goos in the round. Most of them are locked away to you, inaccessible. At first, you only have access to a few easy ones, like an easy Cinema goo, an easy Music goo, an easy Sports goo, et cetera. Each day that goes by, you get to reveal one of these accessible goos and attempt to guess it, time unlimited. Guess it correctly, and you unlock the adjacent goos on the grid, which get a little harder; come back the next day to reveal one of those and guess at it. Guess it incorrectly, and nothing happens, but you will have to get to those adjacent goos by another route.

Here are quick mockups:

1) You have just started, and have access to a few goos.
2) You choose to reveal the goo at B3. You see it on the chart, waiting for your guess.
3) You guess correctly, unlocking the four spaces around it, or...
4) ...you guess incorrectly, making it harder to proceed.

You can reveal as many goos as days in the round have passed. On day 20, you might have been playing each day and now you're on your twentieth goo, or you might just come along now for the first time and guess at twenty goos all at once. It's up to you. Note that you'll still have to unlock adjacent goos in order to reveal that many.

Here's the competition part: The "super goo" is originally a blank slate, impossible to guess. Each time you guess correctly on the grid, you reveal the corresponding portion of the "super goo" image, and you get a hint at the person's identity. For easy goos this will be something unhelpful, like "born in the 1900s." For hard goos, this will be more specific, like "once appeared on this television show." Each hint corresponds to a specific space on the grid, so players who head north will get a different set of hints from those who head south. Players slowly progress across the grid in different orders, revealing different elements of the super goo. The first player who guesses the super goo correctly wins. Each player only gets one guess.

What do you think? It has an element of "randomness" since you might proceed around in different directions from the other players, but not really, since some others are bound to choose the same path you did, and you get to choose whether you reveal an easy goo or a hard goo each step of the way. It isn't timed, but it can still only produce one winner, based on who is good enough to solve that beast of a super goo, which will be quite hard. It rewards people who play consistently, since they have an easier time unlocking goos (dead spaces are tough to get around) and they have a longer time to ponder the super goo than someone who waits until day 49. The system has other advantages, such as busy players being able to play even less frequently than weekly and still keep up.

Here are the downsides that I foresee:

1) Friends might choose to help each other. You explore the left side of the grid while your buddy explores the right side, with a pact to share hints with each other. Normally the nature of the competition provides a natural disincentive for you to help friends - what would you get out of it? - but now there's an incentive to help, since it gives you more hints about the big finish. Is there a way around this? The honor system has proven insufficient.

2) People might still argue that "the hint at D3 was a gimme, no wonder he won" et cetera, since not everybody has access to the same hints all at once. I'll do the best I can to make things fair, but I'm far from perfect. Is there a way around this?

3) Someone might solve the super goo really early, as in during the first week. I'll do my best to make it hard, but it's possible. Is this a problem? I say let 'em have victory if they're that good.

4) There's no way I can program this thing, and create 50 goos, in a week and a half. I would have to delay the next round for another 1-3 weeks. I don't see this as a problem, but you might.

5) Loss comes out of nowhere. I might indicate on the chart who has seen what hint, but whoever solves the super goo will do so very suddenly, ending the round abruptly. Is this a problem?

6) This potentially disrupts the existing goo system, such as numerical goos and dates published and so on, but I'm probably the only one who would care, and I don't. The one big loss might be the "streaker" award, which could become too difficult to calculate.

Are there other problems that I haven't thought of yet? I just thought of this a couple of hours ago, so I'm still turning it over in my mind.

Fyi, once the super goo is solved, I would probably open the grid to all players for the month between rounds, to give you a chance to guess all of the goos for sake of your overall score and for fun.

PS. Smash Bros. fans, I suspect that the "vault" menu may have subconsciously influenced me.

Amy Austin | April 9, 2008
Wow. That sounds pretty cool (and inspired!), actually...

I, for one, am very intrigued...

;-)

Steve West | April 9, 2008
I'm willing to give it a try despite being a creature of habit... I fear change. I'm in.

Justin Woods | April 9, 2008
I'm in for something new, should be interesting though; can't wait...

Scott Hardie | April 9, 2008
Glad to hear it has support so far. :)

Would the "friends helping friends" problem be offset by creating a team and officially playing together? Each of you going in different directions, trying to gather separate clues? How would I separate a winner from among the team (if I did)? And how many goos would I have to make for that system to be playable before I went insane? :)

Steve West | April 9, 2008
This is starting to sound like a GooCon event (which would be cool). An environment set up for team play.

Steve Dunn | April 9, 2008
I think this is a great idea. Let's give it a shot!

I refuse to believe that anyone cheats at the Goo Game. Do they also steal Christmas toys from orphans? Come on. For what? The Amazon gift certificate? No way. Pride is the only reason to excel at the Goo Game, and if you cheat, by definition you are prideless. I don't think it happens.

I am against teams, but if we must, can I be on Amy's team?

Tony Peters | April 9, 2008
hummmm this looks interesting I for one think it's a good direction to take the game in and no I don't mind waiting....Teams would have to vary game to game to make it fair but it might make the game too exclusive

Aaron Shurtleff | April 9, 2008
Sounds like fun fun!!

I'm not really a good team player, but I'll give that a try. ;)

Russ Wilhelm | April 10, 2008
I'm with Steve as far as being against teams, but if it so be, you can still count on me being there, match me up with any group.

My bigger question would be, how are player submitted goo's going to be handled, if at all. Would they be gimme's, or would the player need to work around them? A substitute goo?

Also, what happens if a player guesses incorrectly on all four starter goo's?

No doubt you've already considered these, and have determined how these situations will be handled, but, you know, on the off-chance, I figured I'd put it out there.

In the past, there have been several goo's published that refer to earlier goos by their numbers. I'd say keep them, just don't refer to them during the current round. The publish date could be the same for all 50 goo's.

Scott Hardie | April 10, 2008
Thanks for the input, everybody. I don't think we'll do teams; I was just tossing it out there for discussion.

Instead of one super goo, I think I'll do four super goos, each one taking up one quadrant of an 8x8 grid. That will help the round last longer, and prevent accusations that one person just happens to be familiar with the final celebrity.

I hadn't thought about any of the things you brought up, Russ, so thanks for the insight. If a player guesses wrong on all four beginning goos, I'll probably manually unlock an additional goo to give them a fighting chance, but I find it so unlikely that anyone will miss all four very easy goos off the bat that I probably won't even program for that possibility. We'll see.

As for requests, well, I don't know of a good solution off the top of my head. I could give you that one as a freebie if you reach it, limiting everyone to one request per round – which will probably cause a spike in the number of requests, fine by me. :-)  I could also leave player requests out of the system, but who knows how long it will be until we try something else. And there's also the possibility of a substitute goo, one special goo that takes the place of any player's request when they reveal that goo. What do the rest of you think?

People have been caught cheating in the goo game, both veterans and newbies, though it doesn't happen often. I thought that the tournament structure provided sufficient disincentive to prevent cheating, but it happened in the most recent tournament. I might have no other option but to deal with it the way I always have, restoring fairness the best I can on a case-by-case basis, but I wonder if there's anything we can do with the structure of the proposed "super goo" system to discourage it.

Amy Austin | April 10, 2008
I thought that the tournament structure provided sufficient disincentive to prevent cheating, but it happened in the most recent tournament.

?!!?

Scott Hardie | April 10, 2008
It didn't help the person win, obviously.

Lori Lancaster | April 10, 2008
[hidden by request]

Steve Dunn | April 10, 2008
Huh? Cheating in the most recent tournament? Not just why, but how?

I won because of my wife's pajama pants - fair and square! All you cheaters out there, get your own pajamas!

Scott Hardie | April 10, 2008
Someone outside of the tournament guessed the goo, then gave the answer to someone inside the tournament. If we do have a tournament system again, where only some people are eligible to win but others can still guess at goos, Russ has already suggested the solution: Show the goo to the eligible players first, then only show it to the rest of the player body after they're done competing.

Denise Sawicki | April 10, 2008
Show the goo to the eligible players first, then only show it to the rest of the player body after they're done competing.

Seems fair, I kinda had thought it might work that way already... I know once in a tournament, AFTER I guessed I was curious what was visible to others outside the tournament so I had Darrell log in and look at the goo with his account and nothing was visible. I probably raised your red flags as a potential cheater by doing that. (I was just curious if it was OK to discuss the goo after the day was over or if others were still guessing at it!)

I'm not so interested in winning things really and have a hyperactive guilt sense so I wouldn't cheat myself :-P Add to that the fact that I can't remember when I was last able to get a tournament goo even with INFINITE time to think, having it activated early would have done nothing....

Your idea is cool, I'm with some of the others though in not being a team player. But besides that, I like it :-)

Steve Dunn | April 10, 2008
Someone outside of the tournament guessed the goo, then gave the answer to someone inside the tournament.

How do you know?

And why do you protect the identities of people who do this sort of thing? You never revealed who posted a goo on that rap discussion board, either. Trying to avoid putting asterisks in the record book?

Now is probably a good time to admit, yes, I've been using human growth hormone this year.

Lori Lancaster | April 10, 2008
[hidden by request]

Steve West | April 10, 2008
I got my Golden Imelda after smoking cat tranquilizers. But I never inhaled!!!

Russ Wilhelm | April 11, 2008
How about requested goo's as a freebie space, but has no clue associated with it. That makes the super goo's even.

Limiting to one per round is a good idea for this course.

Honestly, I would rather have to work around it rather than not be able to request a goo.

Scott Hardie | April 11, 2008
I decline to say how I know and who it is. Almost everyone caught cheating asks me not to reveal their identity. If I believed that revealing their identity would prevent more cheating in the future, I would do so.

The two most common methods for cheating are people sharing answers and people creating ghost accounts on the site to test their guesses. The latter is easy to catch, but the former is a lot harder. There are occasionally other methods seen, such as the rap discussion board controversy (someone asked for the answer to a current goo on another web forum by posting the goo itself), and attempts to hack my code, which to my knowledge have been unsuccessful.

Steve West, thank you for the laugh. :-)

Aaron Shurtleff | April 11, 2008
I admit that when I'm stumped on a GOO, I scroll the mouse pointer over it, just on the odd chance that it'll pop up with the answer instead of like GOO1356 (SPORTS). Scott hasn't messed up yet, to my knowledge, but I check! ;)

Steve Dunn | April 11, 2008
Dude... attempts to hack the code??

Whoa.

Lori Lancaster | April 11, 2008
[hidden by request]

Scott Hardie | April 12, 2008
Basic hacking attempts would be things like keeping your guess form open in a browser window after the goo expires and then submitting it with the now-revealed correct answer, which is a simple attempt to circumvent the programming of the site. I'm pretty sure I've set up protections against every sort of violation like that. More advanced attempts would include submitting database queries in the form, which could give you access to all of the data in the database and let you vandalize/erase it if you were malicious. I've put in every protection that I know of against that, but it's not my area of expertise, and there are all kinds of ways to do it. Only one person has told me of his (unsuccessful) attempts to crack my system. I'm lucky that I run a small site, because if this thing took off in popularity, it would only be a matter of time before someone found a way to crack it. :-\

Yes, I am informed of inapporpriate transactions on my site, which are almost always spambots trying to submit Viagra-related feedback, since all of the other forms are behind a login. My feedback form doesn't generate actual email, only a message to me using the site's user-messaging system, so the bots couldn't manipulate it to generate spam to strangers from my server, but that doesn't stop them from trying. I think I'm going to require cookies to access the feedback form; I'll put in a captcha if I have to, but those are annoying.

Scott Hardie | April 14, 2008
After planning this out, I realized it's going to take me even longer than expected. I would have to postpone the new round until June or later. So, let's play one more round the way we have been, and I'll launch the Super Goo system the round after that.

This delay gives me a chance to finish some of the other new features I've been working on, which should come first. Hopefully one will be done tonight.

Allison Bair | April 15, 2008
The new format sounds really cool, Scott. I'm looking forward to trying it.

Chris Lemler | June 9, 2008
How about giving each person a different goo to figure out in the tournament. You could always give someone a different goo to figure out. Instead of giving people the same goo give everyone a different goo. Then after one round see who figured out their goo and advance them to the next round. After the first round is over and all the players guess correctly or incorrectly on the goo you can advance the fastest time to the next round. You could alway give someone else another goo each time. Don't give the same goo to everybody on the same round. Mix up the goos and see who can get it the fastest. At the finals then give the two people that are playing each other the same goo and the person who guesses it right in the fastest time wins.

Samir Mehta | June 10, 2008
[hidden by request]

Samir Mehta | June 10, 2008
[hidden by request]