Troperville

Tools

What's Happening

This is discussion archived from a time before the current discussion method was installed.

From YKTTW

Pika_power: Can someone explain how that "When did they die" row works? I just can't wrap my head around it. Also, surely if it is a Non lethal K.O., they are not even dead? Also, if we see them waking up at the end of an episode with a faux death, surely that means that they are alive?

Earnest: Think about it as a curve. At first they are a lot likelier to turn up alive in the first few episodes or season after a death, but then the odds of revival fade as memory of them lapses. But at 10 years the odds of them returning grow likelier for the sheer shock that no one is expecting it (we use 10, but 3/4 seasons for TV is about right), but any longer than that and they will be so far gone from memory that their return is incredibly unlikely. I don't see where in the algorithm the Nonlethal KO is (it should have been removed a while back) but the Faux Death is under the 1 precisely because this is the least dead a character can be. However, if you don't thin it should be there, just elaborate some where and we can get it off there (heaven knows the algorithm has seen a lot of deserved changes).

Pika_power: I think that the "When did they die" row should be reworked. We have in adjacent blocks "5 and 10+ years ago" and "10 years". That is difficult to understand, assuming it makes sense (which I still don't see). The Faux Death should also be removed, because it's a confirmation of whether they are dead or alive, not just an indication. Short of them waking up in hell/heaven, I can't think of any situations where a character waking up at the end of an episode doesn't mean they are alive. Perhaps replace it with How You Dare Die On Me/Please Wake Up or something else that conveys someone crying over their body saying "I still need you..." Any suggestions? Also, I think we should add a row for the character's standing with fans. Highest would be The Scrappy lowest would be Ensemble Darkhorse. Which still has Shirley as a four.

Zeke: I love this. But "Next!" could be a little clearer... and may I suggest the name Sorting Matrix Of Deadness?

Kizor: While I'm off hatching my dastardly plot, someone who has the skillz should integrate Really Dead Montage and possibly How to Kill a Character.

Earnest: My cleverness fails me on the How to Kill a Character front, but Really Dead Montage has opened up a delightful category to explore. As to the name, it's really all the same, the idea is you "add up" the circumstances in each column for a rough "death quotient", actually, that might be fun to put in... ^_^ (Also, replaced "Next!" with "Deader than disco", but feel free to funny it up.)

Zeke: The column titles are much improved — good job. The article name is obviously pretty unimportant, I just think Matrix sounds cooler. And the "how to use" should definitely be in there (especially if we're gonna call it an algorithm... can you tell I'm a math grad?).

Jonny D: I've gone ahead and added a redirect from Sorting Matrix Of Deadness, so now both can be used

Ununnilium: I'd switch which one's the "real" name and which one's the redirect, but other ise, good stuff.

Earnest: I added an example of the death of supes and a prediction for the return of captain america based off of trope factors, but I was wondering if adding "Hard math" is worthwhile. How would you go about adding math to it? Assign a number value to each column, add up the rows and divide by the number of applicable categories? With higher/lower scores representing the "half-life" (or half-death?) before the character can come back.
  • Mike: No, I don't think we can have a rule applied to every column. Is a Dead and Buried character in a show where Anyone Can Die who died from a bridge to the skull MORE or LESS likely to come back than someone who just got shot in the heart in a show where Anyone Can Die? As far as I can tell, each one of those makes the character *less* likely to come back, so it should be multiplied. And yet, if we go over to the right-hand column, each one of those added makes a character much more likely to come back, so there it should be division. Or maybe there is a way to do this, but it would probably be more complicated. What if you look at every circumstance that applies, and then multiply them all together (So, a protagonist (1) who died in a Heroic Sacrifice (3) and got a Really Dead Montage (4) would have a score of 1 * 3 * 4 = 12. If its a show where anyone can die, it would be 1 * 3 * 4 * 4 = 48. If its nothing but 1's, we get a total of 1. That's a hefty growth rate, so maybe we should get some way to curve them down to lower values. Maybe log (base score), which would give us 1.079 and 1.68 respectively. Or, to get something just a bit bigger, ln(base score), which would give 2.48 and 3.87. If there's nothing but "1"s in the equation, we get a Really Dead Score of 0 for both base-10 logs and natural logs. Or maybe, if we don't like logarithmic scales, just go with the square root (giving 6.93, 3.46, and 1). I don't know, I'm rambling, its late and I took my sleep meds a while ago. However, what we've currently got just doesn't work.

Earnest: Well, my math is very rusty, so that may be why it seems overly complicated to me. I lack the expertise to test out your proposed methods, but we should keep in mind that the scale on the trope is not scientifically rigorous, and dramatic license can and does trump it. It's a general rule that can always be made more exact, but before we change the way we evaluate the deadness of a character we'd need to find a way to seriously evaluate the merits of all these different methods in order to choose between them. In the end, I just worry this would likely discourage people from adding examples if the most accurate method is overly complicated. I don't mean to be discouraging, but we should remember that TV Tropes should be both fun and informative, not too Serious Business.
  • Mike: I wasn't being too serious, and I misread the math (or it is being implemented in a slightly different way). A simple average of the scores works fairly well. I somehow (probably because it was about one in the morning and the sleep meds were starting to kick in) read the math as (column value / number of cells in that column that apply) for each column, instead of ((column value of cell A that applies) + (column value of cell B that applies) ... etc)/number of cells that apply. I also somehow (probably because I'm a little bit crazy) hit upon the idea of a log scale. I don't think my explanation makes much sense, so here goes:
What seems to be being used (straight average)
Dead and Buried (4), Anyone can die (4)
(4 + 4)/2 = 4 (okay)
How I misinterpreted it
Dead and Buried, Anyone can die (so that's two entries from the "4" column)...
4 / 2 = 2 (somewhat less okay)
My crazy ideas
Dead and Buried (4), Anyone can die (4)
Version A: natural logs
ln (4 * 4) = ln(16) = 2.77
Version B: log base 10
log (4 * 4) = log(16) = 1.204
Version C: square roots
sqrt(4 * 4) = sqrt(16) = 4
(if, say, the dead character was a Recurring Character, adding a "2" to the equation): sqrt(4 * 4 * 2) = sqrt(32) = 5.65
  • (comment continued from above mathy part) so, my idea is probably unnecessary. However, because the idea of using logarithms on TV Tropes amuses me, I support Version A. But...anything involving logs is almost certainly overcomplicated for a site like this.


i would just like to point out that disco is not dead... disco is life
Lale: Could we start a specific page for examinations of individual cases like the Superman and Captain America paragraphs? Just for Fun... and what a lot of fun that would be.

Earnest: Behold! The Wild Mass Guessing Sorting Algorithm of Deadness! Hmmm... maybe I should have used a different title to differentiate it a bit?

Lale: That was quick :)

Earnest: Well, I'm bored, it was easy to do, and it's a good idea. ;)

Fast Eddie: The "when" row is not making much sense. Any objections to chopping it? // Later: Wait! I get it, now. My bad.

Earnest: Would it help if the question were changed to "How long ago did they die?"?

Fast Eddie: Nah. It is perfectly plain. The thing that threw me was that there were no links out to the column items. I was reading it as a crafty index into Death Tropes. All me, all in my own little brainpan.
Earnest: Prior to the crash, we'd agreed to store the Video Game category here for later use, as it didn't really apply to how dead characters are, even inside of games.

Video Games Plotline Death Final Death Only Mostly Dead Non-Lethal K.O.


Ironeye: There's been some talk in WMG and the corresponding discussion page about the use of Faking the Dead and Not Quite Dead. Considering that these two tropes only apply if the audience knows the character is alive, they don't really fit on a chart that deals with dead characters. I suppose it works if the character in question is known for one of these two tropes (such as Homer's mother for the former or Apophis for the latter), but if so, we should probably make a note of it somewhere so that a character can get a 2 in "Body Found?" without another argument starting. This same sort of issue could come up for Faux Death and Waking Up, with the exception being works known to screw with our heads by throwing in a Dying Dream or two.

Earnest: I've moved things around a bit. Faking the Dead and Not Quite Dead are now in "Died" and come back? I felt it appropriate since many times there's no hocus pocus involved, just faking having died. To fill the hole this left in Body Found? I added Human Popsicle, as frozen characters are effectively Put on a Bus and can be revived quite easily.

Dukia: "Dead and Buried" can have no guarantee on a character's likelihood of coming back from the dead in horror films, particularly vampire and zombie ones, so I think a note should be made there. Just don't know how to phrase the note about that one not decreasing the likelihood of a character's resurrection in a vampire/zombie movie without making the square way too huge.

Earnest: I think the best bet there is that we add a bit to the main description that the algorithm covers returns to full life, and not undeath, which is equivalent to the First Law of Resurrection.