Mar 18 02:49
@ToddWilcox Well, if we can't agree on "there is an ordering", I don't suppose anything I can say will convince you. In that case, let's just agree to disagree. Regardless, thanks for your time, and have a nice day.
Mar 18 01:20
Results generally hover around a 20% chance for women to make up half or more of the top 10% of the list, which while possible, is clearly not equal.
Mar 18 01:19
That being said, I've again biased the simulation in favor of women. This is the Fisher-Yates shuffle algorithm, which you can think of as repeated random insertion. In other words, it picks a random person and ranks them first, then a second random person to rank second and so on. I've biased it such that for each decision, three attempts are made to draw a woman first.
Mar 18 01:15
from random import randint
from math import ceil

#Fisher-Yates shuffle algorithm, but biased so that it attempts to draw f multiple times
def biasShuffle(population, attempts):

for i in range(len(population)):
#attempt to draw f multiple times - break if found
for j in range(attempts):
swapIndex = randint(i, len(population)-1)
if population[swapIndex]=='f': break
#swap array elements
population[i], population[swapIndex] = population[swapIndex], population[i]

#parameters ----------------------------------------------------------------------------
Mar 18 01:14
Third, here is a second variant of the simulation, that doesn't rely on any fake scores or distributions at all. This one simply creates an array of men and women, and shuffles them randomly. Surely you at least agree that "an ordering exists" is a reasonable assumption?
Mar 18 01:13
@ToddWilcox Ok, I've read your comments a bit closer and did some research - you're talking about distributions as in uniform/normal/Gaussian and so on, right? In that case I've got a few questions:

1. First, pretend we weren't talking about academic presentations for a moment. I'd like to know what explanation you have for rankings in chess that doesn't rely on the gender proportions. Remember, the point I'm trying to prove is that proportions within the population are a predicting factor (though not the only one) in statements about subsets.
Mar 17 03:36
*correction: typo, one too many zeroes. Top 100, not 1000
Mar 17 03:11
@ToddWilcox Ok, forget about the simulation then, since that seems unconvincing to you. Then how about a real world example? How do you explain the ranking bias in the top levels of chess? For context, about 10% of FIDE players are female. Of these, there are less than five in the top 1000 players of the open league. This is a field where we can definitely rank people by ELO.
Mar 16 23:14
@ToddWilcox But... the magnitude and bounds of the score don't matter in the slightest? If you want scores between 95 and 100, then just pass my arbitrary score through f(x) = 0.05x + 95. 0-100 is just a standardised scale for demonstrative purposes.

If you want to relate it to something in the real world, then you could go off exam scores, or oral presentation scores (since everyone in the department is likely to have gone through a university education).

But also, yes, it is heavily biased in your favor. The men have not been extended any such courtesy, and their scores range between 0
Mar 16 21:50
Third, I never said presenters are randomly selected. In fact, the simulation code is selecting the top 100 people as 'presenters'. The random selection is of the scores of the population (which again, I remind you have been heavily biased in your favor).
Mar 16 21:48
The 4:1 ratio is admittedly arbitrary, but a quick online search shows this to be very similar to what you'll find in engineering (about 20.9% women).
Mar 16 21:47
@ToddWilcox Well, now you're moving the goalposts, aren't you? Quote, "It’s possible that there are more capable men in the field than women, but we cannot conclude that merely because of there being more men." I'm demonstrating that some knowledge about the distribution of gender is valuable information in determining the distribution of highly ranked people. Also, the whole point of sharing my code is so that you can experiment with the numbers and come to your own conclusions.
Mar 16 00:02
(Code pictured with indentation because chat can't do that)
Mar 16 00:02
Mar 15 23:55
I'm no statistician, but 0 out of 1000 trials when the experiment is heavily biased toward women having high scores is pretty convincing, wouldn't you say?
Mar 15 23:54
Script below:

from random import randint

#edit this to try more times
trials = 1000
successes = 0

for trial in range(trials):

#reset results
results = []

#create simulated test scores
for i in range(1000):
#edit this to change proportion of men to women. Currently set to 800 men and 200 women
if i<800:
#edit these to change scoring bounds for each gender. Currently set to 0-100 for men, 50-100 for women
results.append((randint(0,100),"m"))
else:
results.append((randint(50,100),"f"))

#sort result by score
Mar 15 23:53
(didn't realise pressing enter sends a message)

3. 1000 trials of the simulation.

I have run this script a few times, and every single time it has returned 0 successes (a success being defined as more than 50 women in the top 100).
Mar 15 23:52
@ToddWilcox No, we absolutely can. Here, I wrote a Python script to simulate the situation. The current parameters of the script are:

1. 1000 people, of which 200 are women.
2. Scores between 0 and 100. I have biased the simulation in your favor by giving women a score between 50 and 100.
Mar 14 05:38
*Correction: Also add rank 26 to the list. I was short by one.
Mar 14 05:37
Also, sure, you can argue other factors are present which could possibly make the average female candidate rank higher, and that should warrant an overrepresentation of female speakers. However, in OP's question, they have noted symptoms of this overrepresentation being pushed too far.
Mar 14 05:31
@ToddWilcox Maybe this explanation will make sense. Let's assume 100 researchers, of which 20 are female. Assuming equal talent, you would expect that, if you ranked all 100 by speaking ability, the women would be equally distributed across the ranking. So let's take an idealised ranking, and assume rank 1 is a woman, rank 2-5 are men, rank 6 is a woman, 7-10 are men, and repeat.

Now, if you take the top 10 of the absolute ranking and choose these to be speakers, you get eight men and two women.
 
Jul 4, 2024 09:30
(I'm not Christian, this is not my argument, but) this common response for "Eternal torment for finite crimes" makes sense to me: The severity of punishment is dependent on the standing of the afflicted. For example, slapping a child is frowned upon. Slapping a parent warrants a grounding. Slapping the Queen of England will probably land you in jail. As such, sins, which offend the mightiest being in all creation, are deserving of a commensurate eternal punishment.
 
May 12, 2024 09:17
@tnknepp Let's assume for the sake of argument that before the Big Bang there was nothing. Then one of two things is true: either the universe came into being without a cause, or a higher entity caused the universe to come into being. The first possibility would have me believe things can exist without cause, which admittedly feels improbable. But the second possibility would have me believe a higher entity can exist without cause, and willed the universe into existence. Aren't these pretty much the same? To me, this is an unconvincing argument for God as it only layers on another question.
 
Mar 4, 2024 05:46
@oAlt Yep, that's it, nice one. Making cryptics seems to be pretty fun, maybe I'll try some other wordplay formats
Feb 29, 2024 23:09
Another try at a cryptic:
Container and shortened gratitude is protection (6)
Feb 23, 2024 23:54
Yep that's it, and thanks for the tip :)
I'll give it another go soon
Feb 23, 2024 01:34
Just read the guide on making cryptic clues, here's an attempt at one:

Soda and dog toy is great (9)
Feb 8, 2024 22:42
Oh right, so [letter]+[alpha characters] is too recognisable/not obfuscated enough? Ok yep I get what you mean now. Again, I guess just down to not knowing what's common knowledge
Feb 8, 2024 22:40
@SnySmartie That's the point - the cipher looks like a jumble of random characters until you try the partial key, at which point the pattern of U+ helps you reverse engineer the rest of it
Feb 8, 2024 22:38
@codewarrior Not sure what you mean by this though. In my mind the partial key could be used to get some of the characters of the plaintext and hopefully show the pattern of 'U+' at the start of each group
Feb 8, 2024 22:35
@codewarrior Thanks for the feedback. I've seen people on this site pull off things I thought were far more impressive, but I suppose being new to puzzle-making I'm not too aware of what's commonly known by the community.
Feb 8, 2024 05:04
Ok, here we go then

https://puzzling.stackexchange.com/questions/119750/some-of-the-clues-seem-to-be-missing

I've written an intended answer as well
Feb 8, 2024 03:21
Hello, is it ok to ask for puzzle design feedback here? I made an enigmatic puzzle a while back but it didn't get much attention
 
Jan 12, 2024 10:09
@user76284 Well, why didn't you lead with that? Your original question has no value under total utilitarianism, so now we've just wasted time bickering over semantics, instead of moving onto issues with total utilitarianism itself. Also, why did you post this here instead of chat?
Jan 12, 2024 10:09
@DKNguyen Sure, but I find that to be less compelling. In the case where no choice exists but to kill a group of people, common sense aligns with utilitarian thinking in consigning the smallest possible group to death.
Jan 12, 2024 10:09
@NotThatGuy What if I tweak the scenario a little? From reading news about climate change, it's easy to see that humanity tends to leave important decisions to the last moment. What if global food supplies are dwindling, and a gradual population adjustment won't make it in time? As far as I can see, this situation has only two choices - genocide or widespread starvation.
Jan 12, 2024 10:09
@NotThatGuy I'm an advocate of utilitarianism for non-extremes, but to play the devil's advocate: How about the case of overpopulation? In a possible future where Earth doesn't have the resources to support the human population, a valid utilitarian summation would involve a Thanos-like genocide of large swathes of people. The thinking being that without said genocide, society as a whole could collapse, leading to an even larger number of deaths.
 
Jan 11, 2024 22:30
@ajd138 Would you care to respond to my first point? To reiterate, there is a simple utilitarian explanation for being unhappy in a society where people are randomly murdered. Even if painless (most murders aren't), the victim of the murder loses all future utility they may have experienced. In my opinion, that's sufficient cause for widespread fear and doubt.
Jan 11, 2024 22:22
@user76284 Well hold on a moment here. I've just noticed that your measure of 'average happiness' isn't correct. The whole point of utilitarianism is a sum of happiness, not an average. Otherwise, how would the trolley problem make sense? Under the measure of average happiness, the distribution of happiness would be important in making the decision, but at least in the basic premise I guarantee every utilitarian would choose to kill one to save five.
Jan 11, 2024 11:28
@user76284 I do admit that under this interpretation I fall headlong into the mere addition paradox. I haven't got a good response for that.
Jan 11, 2024 11:28
@user76284 Is that not exactly what I've defined? Over the course of some event, the agents present have their level of happiness/pleasure altered by some amount. My interpretation states that the action that maximizes the sum of these changes is the moral choice. Also, while I'm here, allow me to address the issue of "could". I'm stating that under utilitarian thinking, there are choices that are better than suicide, and as such suicide is not obligatory.
Jan 11, 2024 11:28
@user76284 I think we're interpreting 'utility' as different things. Correct me if this is wrong, but your comment implies that utility is a measure of how happy a person is at any particular moment. My interpretation is more interaction-based; ie. A certain event creates some amount of utility, which is a function of the agents' levels of happiness before and after said interaction occurs. Hopefully this helps to clarify my last answer?
Jan 11, 2024 11:28
@user76284 The other guy is making a fool of himself, but he did also answer the question while doing so. I'll reiterate and add some of my own thoughts. The most moral choice is not suicide, because that person could spend the rest of their life doing, say, volunteering at a soup kitchen, or making/donating money to charity. If said person is an evil figurehead (such as Hitler), then their eventual natural death should provide the same utility as a premature suicide. Therefore, no, I don't believe any such obligation exists.
Jan 11, 2024 11:28
I don't think this is a good example. In this scenario, there is something you've left out of the calculation: the sum of all happiness the victim would experience for the rest of their life, were they not murdered. I don't think there would be many circumstances where the pleasure gained by the murderer would outweigh this.