897 karma · joined November 2, 2013
I hardly use it, though, because I usually have < 100 open tabs, not thousands like others have. I identify tabs by their tree structures (parents, children, siblings tabs) and the prefixes of the titles, whose lengths don't depend on how many tabs opened, because the tabs are arranged vertically.
I've been staying with Firefox not for the performance (today Chrome loads Google sites like YouTube faster), but mainly for Tree Style Tab extension. I couldn't imagine opening more than a dozen of tabs without it.
But when you really think about it, moving on is as pointless as feeling dreadful. In fact, anything is as pointless as anything else. It might seem like an insane question, but why exactly is moving on somehow more desirable than feeling dreadful, if everything, including the universe itself, will be gone?
Life is about the journey, not the end, many people would argue. But in reality, most people live their life with long-term thinking. People study hard, so they can get a good job. Then they work hard, so they can retire comfortably. There's always delayed gratification in most life stages and even life activities (saving, exercising, dieting).
If life were all about the journey, we wouldn't need to feel pity for those who "wasted" their life, as long as they enjoyed it. Being drug addicts and die at twenties would not be "less" then being a president, they would be just.. different. But no, most of us think they are not just different, and it says something about human beings.
Maybe I was not explaining it clearly enough, but I agreed with this. Similarly to what you said here, I said in the original post that to survive and to reproduce is more of a tautology than a purpose. What I meant is that if there must be a purpose of life, survival would be the closest thing to it, but in some strict senses of "purpose", it is not a purpose.
What's more interesting is the second implication, which suggests that there would be no objective morality if the three premises were true. I wonder if you agree that there's an objective morality. Or, to put it into a question, how do you think humans, almost universally, can distinguish good from bad?
Wow. By the fact? Really? I thought there's a small chance that the discussion with you could be saved, but now you've said this, no, this discussion won't get anywhere, ever.
What you've been doing is coming up with an "objective" moral compass / objective function of life (maximizing happiness, minimizing suffering) that you strongly believe literally everyone will agree in such and such conditions. No, they won't.
You just keep arguing against what I didn't support or even mention. You were focusing on nonessential details of imperfect examples, which should at least give a glimpse what I've been meaning all this time. Or maybe you really didn't understand of what my counter-arguments really are, after all.
> No, we absolutely do not all agree on this ... Nor do many people, especially pacifists, agree that soldiers are automatically good people.
> ...
> The answer is: it depends on the context. It depends on the people.
The fact that we—and people in general, e.g. pacifists you mentioned—don't agree what's moral or not just proves my point that it's probably impossible to define objective morality from physicalists' point of view. As I've said many times, the closest thing to objective morality is a concept of good or bad based on whether it's beneficial for survival or not. But even this, as I explained in other comments, has its own problems.
> The purpose comes from why the creator created it.
Some higher being might impulsively created the universe out of boredom, in which case there would be no ultimate purpose at all. But maybe they created it intentionally, with a purpose, which is the ultimate purpose of universe and everything inside it.
So no, it's not what you condescendingly said "wordplay at best." There can be a relationship between the purpose of a creation and the reason why the creator created it. In fact, for many everyday things, the purpose of their existence is exactly the reason why the inventors created them.
(I'm not arguing that there exists a creator of the universe. I'm just showing how irrelevant, to put it mildly, your seemingly strong argument is to the post you replied to.)
> I think I need to emphasize a point better: there are infinitely many objective measures.
No one here or anywhere is arguing that there are no objective measures. Of course, there is! The question is whether there is an objective measure from which good or bad is determined. Or in a logical statement, whether this statement is true: There exists something that is an objective measure AND from it good or bad is determined.
> It's objectively true that humans ... It's objectively true that humans ...
There are humans, albeit in minority, that take pleasure from painful experiences. While most of us want to maximize our lifespan, soldiers do not. They consider protecting their nations are far more important than their life. While most of us want to minimize suffering, Mother Teresa was instead experiencing prolong suffering for the sake of, not her family, not her friends, but strangers. Despite all of this, we all agree that what soldiers and Mother Terese did is "good." So no, none of your "It's objectively true that humans ..." statements are true, unless, as I said in the previous comment, you're willing to redefine "objective" as the majority's preference.
> We should seek to perform those actions which minimize the number of rusty nails jammed into eyes and maximize the number of times humans can enjoy nice meals when they're hungry.
I assume this is the objective measure that you think from which good or bad is determined? Say all humans agree with this objective, the obvious follow-up question could be: Which is worse: having 10 nails in each pair of eyes of 10 people, or having 1 nail in either eye of 100 people? Even if everyone agreed with the initial objective, they wouldn't unanimously say one case is worse than the other.
Okay, I suppose what you wanted to say is to minimize suffering and maximize happiness, which is a very common argument. It's a seemingly crazy question, but why? Pain is not inherently bad. If you think about it, pain is just a signal to the brain that something dangerous is happening. From the survival's perspective, it's even good! Unnecessary pain is bad, you might say. Not really, because it doesn't change anything. Only when you argue that excessive pain can be harmful to survival that the question of why can end. Why maximizing the chance of survival? Otherwise, life couldn't survive and we wouldn't have this discussion. Yes, it's a tautology. In fact, I already explained this in my original post [1].
As I explained in my original post [1], if the three premises written in the very beginning were true, "good" or "bad" would be nothing else other than being beneficial for survival or not. But this view of morality is not without its problems. At the end of the post I also mentioned that morality doesn't always align with beneficialness for survival. So, what now? I have written a follow-up in [2] as a response to another commenter.
---
> To be honest I'm now a bit confused on your position, as you explicitly brought up points in your previous post that elucidated some of the cracks in attempting to treat survival as the "one true goal": > ... > perhaps first I should ask how this line of questioning doesn't lead you away from survival as the ultimate objective function.
To be clear, I don't hold any strong position. All my previous comments are based on the assumption that the three premises I wrote in the very beginning of the first post are true, but I'm not arguing that those premises are true. I was just elaborating the implications if they were true:
1) To survive and to reproduce would be the closest thing to the ultimate purpose of life.
2) What traits we consider "good" or "bad" would actually be just what traits are beneficial for survival or not.
When at the end of the first post I brought up the fact that morality doesn't always align with beneficialness for survival, I was trying to do reductio ad absurdum.
Maybe the three premises were not all true after all. Maybe there's something beyond the physical universe*, from which some higher being impulsively created the universe out of boredom, in which case there would still be no ultimate purpose. Or the higher being created it intentionally and with some purpose.
This might also elucidate the common argument, originated by C. S. Lewis, that all human beings know something good or bad because God imbued humans that ability. A common misconception interpreting this argument is to say only theists can do something good. Knowing is not the same as doing. All humans have the ability to distinguish good from bad, regardless of what they believe in, or whether they want to do something good or not.
---
* We don't even know if the observable universe is all physical. No one knows what dark energy is, maybe it's not energy after all. In fact, dark energy is just a placeholder—we might as well name it XYZ.
I actually agreed about this. As I said in the first comment, to survive and to reproduce is more of a tautology than a purpose. If the three premises I mentioned in the first comment were true (I'm not arguing they are true), then to survive and to reproduce would be the closest thing to the ultimate purpose of life.
> Ultimately, meaning and purpose are what you make them.
While many, if not most, of us have some purposes of our own life, that doesn't change the fact that life as a whole would probably not have an ultimate purpose. Whether it matters or not is another discussion.
This is a very common argument that comes from a misconception of what theists argue. There's a difference between knowing what's good or bad, and doing it. What theists argue is not that only who believes in God can do something good. They argue all human beings know something good or bad because God imbued humans that ability, regardless of what they believe in, or whether they want to do it or not.
Interestingly, Christianity is unique regarding this matter. While most, if not all, religions say you have to earn it in order to go to the desirable afterlife (heaven/paradise/etc.), Christianity says no one, whatever good things they do, can earn it. Only because of God's grace they can be saved. Also, we love because God loved us first. We do good things not for earning an eternal life, but for an act of gratitude that God has saved us.
First of all, who are "we"? Second of all, you just proved my point, unless you want to redefine "objective" as something that the majority accept. Also, why are those properties desirable? Because the majority prefer it? There would be no objective good or bad, then—it's just the preference of the majority.
But my whole point is that there can only be one ultimate objective function: to maximize the chance of survival. There might be some other objective functions derived from the ultimate one, but none can be standalone.
Other standalone "objective functions" can always be asked "Why is it desirable?" For example, why is minimizing suffering desirable? Pain is just a signal to the brain that something dangerous is happening. Murder is "bad" because a species that has fewer counts has a lesser chance to survive. Torture is not necessarily "bad", unless it's argued that it in the long run will lessen the chance of survival.
It's actually no more of a purpose than a tautology. It just happens that a species that survives and reproduces will continue to live.
That also means that there's no good or bad. There are only what traits are advantageous to survival and what traits aren't.
For example, being trustworthy is "good" because trust within a group is a prerequisite for achieving a relatively advanced society like Homo Sapiens has built. Breaking into someone's house is "bad" because otherwise we would only be as advanced as other primates that take each other's territories all the time.
But when you think about it, there are "good" things that aren't advantageous (or even are harmful) to survival, and there are things advantageous to survival that are considered "bad" by most people. Many examples would be highly controversial, so I'd pick a less controversial example, but probably not the best one: eugenics.
So, how exactly do we know which is good, which is bad? Is there even an objective good or bad without invoking the idea that some higher beings say so?
So, unless you want to permanently move your PC (and lose your desk setup with a monitor, a keyboard, a mouse) OR to have another PC dedicated for your TV setup in a living room, then no, you can't play on PC while sitting a couch.
That's a new kind of insult LOL. In my case, the slimmer side of PS5 barely fits my setup in a small room. Maybe my case is uncommon, but Xbox Series X won't fit my setup.
In the context of seq2seq models, attention is a technique to compute a weighted average of hidden states of the encoder. When I first realized this simple fact, everything finally clicked.
In constrast to a vanilla seq2seq which takes only the encoder's hidden state at the last timestep as the context vector, the context vector of a seq2seq with attention is a weighted average of all hidden states of every timestep. The weight of a hidden state of the encoder is a similarity score between the hidden state and the decoder's previous output (the decoder's current state). The similarity function can be as simple as a dot product, but there are various ways to do it.
Attention can improve a seq2seq model because now the encoder's last hidden state doesn't need to well represent the whole input sequence, which is hard if the length is long — now the decoder takes at every timestep all the encoder's hidden states and computes an average of them with the weights uniquely different from the other timesteps. The weights at a timestep represent which input words are more important and thus to focus on when the decoder is to output a word at that timestep.
More generally, attention takes a set of values vectors and a query vector and computes a weighted average (or more generally, a weighted sum, if the weights don't sum up to 1) of the values based on the query. In the context of seq2seq models, the values are the encoder's hidden states and query is the decoder's previous output.
Which is very very unlikely. When you don't know the answer, you'll guess, which is like flipping a coin.
Suppose you have zero knowledge and always guess the answer. Can you get unlucky that you get more wrong answers than correct answers? Sure. But the more questions you get, the closer your accuracy is to 50%.
Another way to look at it is to realize that a binary prediction has the highest entropy (uncertainty) when it splits fifty-fifty.
When the task is choosing between two choices (in this case: AI/Human), the worst you can do in average is not 0% correct, but 50%, which is a coin flip. If a model—whether it's an ML one or is inside human's mind—achieves 40% accuracy in a binary prediction, it can increases the accuracy to 60% by just flipping the answers.
The more interesting numbers are precision and recall, or even better, a confusion matrix. It might turn out that the false AI score and the false human score (in the sense of false positive/negative) differ significantly. That would be a more interesting report.
I wonder how many off-by-one bugs have happened undetected because a slice is unintentionally resliced beyond its length. Instead of crashing, so that the issue is known early, the program will still run with inconsistent data.
It's just a tautology. If you understand something, of course by definition you aren't confuse about them. By using the same logic, all languages have "very little magic."