They don't just "sound" reasonable, they *are* being reasonable.
We have never had any economic models with any level of accuracy, ever, in the history of mankind. Asserting that you know "how damaging" *anything* could be to the economy, 80 years ahead of time, is wishful thinking at best.
It might not be satisfying to live with uncertainty, but that *is* the reasonable position - we don't know if increased CO2 emissions and any temperature increase due to that is going to be a net positive, or a net negative. Anyone who tells you otherwise has made a leap of faith.
Not a troll, but yeah, this is going to be seen that way.
Once you've corrupted academia, and given degrees to people based on the color of their skin, or their SJW credentials, can you use degrees as a reasonable proxy for skills anymore?
Sounds like the meritocracy is going to work its way around attempts to thwart it.
Have they also asked for a non-censored search engine in the United States?
It seems to me that Google's activity in China is simply an expansion of their current US censorship, rather than anything new. Sure, the keywords might be "falun gong" instead of "conservative news", but it's silly to protest their activity in China without acknowledging their activity in the States.
Insulin doesn't drive fat accumulation without excess consumption.
I'm agreeing with you as hard as I can.
I think what you don't see is that excess consumption doesn't drive fat accumulation without insulin:)
Even further, I think what's interesting is that when you see a 500 pound fat man eating pizza like he's starving, it's because his muscles *are* starving, and his fat cells, under the influence of insulin, are stealing all the energy from his muscles.
Now, is it "excess consumption" if your muscles are starving? If your energy is being partitioned poorly into fat cells, it's actually a *survival* skill to eat more to provide your muscles energy:)
Take an aluminum rod, 2 feet in length. Place one end of the rod in a pot of warm water. Measure the time it takes for the other end of the rod to increase in temperature and reach equilibrium.
Now, do the same experiment with a human. Place one hand in a pot of warm water. Measure the time it takes for the other hand to increase in temperature and reach equilibrium.
Now, of course, basic physics still works - but what you're really interested in isn't "calories in/calories out" as measured by "food in the mouth, physical activity of the muscles". You could put a metal ball with 10,000 calories of energy in your body, and not be able to metabolize any of it. So the *quality* of the calorie counts too.
In the case of insulin, which drives fat accumulation, it is driven by blood sugar levels, which is driven by carbohydrate intake (particularly refined carbohydrates). If you don't include that in your calculation, you're using a simple formula for a complex scenario.
That's basically saying you'd avoid SEO abuse by doing something about it which is not even slightly a proposal as to how.
Are you saying that it is impossible to program algorithms against SEO abuse without engaging in political viewpoint discrimination?
You keep conflating things - I'm not sure if you're engaging here in an attempt to understand a point of view contrary to your own, or just get a quick endorphin rush from arguing with someone on the internet:)
If I want to find about about homosexuality in various branches of tetrapoda, I don't want to be deluged with weird ranty and utterly wrong "opinion pieces" from idiots playing politics.
So, therein lies the rub I suppose - how do you propose to create a search engine that reads people's minds? Someone with the same search terms might be looking for something completely different than you are.
Now, maybe it's as simple as designing pre-calculated "safe-search"-esque modes where someone can "filter-out-right" or "filter-out-left" or "filter-out-sarcasm" or "filter-out-humor" modes. But assuming you can build a single mode that will in fact, satisfy the various intents of people who are resistant to mind reading...I'm not sure if that's helpful.
Well I geneally want my searches to provide useful information, so why would I want it to persistently return utterly uselsss junk?
Why should you trust anyone to decide what is useful, and what is useless to you? Why should anyone trust you to make that decision for others?
We haven't. I just searched for "abortion". I got, in order:
The NHS
Funny, google puts a planned parenthood link on the top of my results:
In fact, on that first page, not a single anti-abortion link comes up.
Now, before you freak out, please, understand this - I'm firmly pro-choice. I buy the rapist Bill Clinton's formulation of "safe, legal, and rare", and I'm not by any stretch of the imagination a pro-life protester. But this kind of censorship that Google does, in the United States, against political points of view, is brutally obvious.
I mean, brutally obvious.
Now, it may very well be that your search, done in the UK, is less censored that US results...but whatever your opinion on the matter, we should be incredibly reluctant to give powerful gatekeepers the ability to drive the conversation through editorial bias eliminating points of view from the material being presented. Any bit of joy you have over someone like Alex Jones being downranked is going to come back and bite you in the ass when a different political bias gains power, and works against your interests.
So go on do tell, how do you think one could make a search engine which is utterly "unbiased" and yet not prone to the trivial gaming of the sort that was rampant in the late '90s.
Well, I would make whatever algorithms being used transparent - obviously trivial stuff like keyword abuse recognition, and other metadata analysis that would indicate SEO abuse - however, I wouldn't allow for point of view discrimination. For those people worried that their view point was being discriminated against, they'd have the algorithm to look at, to double check that it isn't simply taking some blacklist of conservative celebrities, and shadow banning them.
The problem here is that the Overton window keeps getting thinner and thinner, and pushed further and further to the left. This isn't healthy or responsible.
Just about everything. But then again I don't consider not returning an Alex Jones rant about turning frogs gay when one searches for "anphibians" to be censorship.
So, you don't have a problem with returning an Alex Jones rant, when you search for "gay amphibians"? Or returning a wattsupwiththat.com article when searching for "global warming"? Or autocompleting "hillary clinton is...married to a rapist" if that is in fact the most common completion?
My fear here is that we've gone beyond "someone searches for amphibians and gets Alex Jones", and gone into "someone searches for abortion and only gets pro-choice views". Deciding what is "useful" for someone a priori seems like something we should be much more careful about.
Okay, thank you for responding to two of the lines.
Can I assume you agree with the third one?
"How about giving users the option to "opt-in" to a politically correct blacklist/whitelist, and allowing the default to give the raw results (sort of like "safe-search")?"
Only if you have a stupid definition of "censor" that doesn't match anything else. And even then only if you studiously ignore history.
I'm not sure if anything would count as censorship by your definition, other than "well, censorship is when something I personally like is restricted".
Is there any content that you find distasteful, offensive, misguided, or otherwise incorrect that you *wouldn't* censor?
I'll start with my example - I think the bible is distasteful, offensive, misguided, and generally incorrect. I do not think any of it should be censored from search results. Can you come up with anything like that, so we can understand what you might *not* censor?
The front page contains what, 20 items? How do you propose that google or indeed anyone present you with 20 items only from your search terms without as call it "censoring".
How about just having a content neutral algorithm that does a text search without any manual tweaking of the results in order to get a pre-ordained result?
How about giving users the option to "opt-in" to a politically correct blacklist/whitelist, and allowing the default to give the raw results (sort of like "safe-search")?
There are lots of ways of providing a search engine without being censorious.
"Article 35. Citizens of the People's Republic of China enjoy freedom of speech, of the press, of assembly, of association, of procession and of demonstration."
That is no more “hypocritical” than distinguishing between murder and self-defense
You're falling down a deep semantic hole there:)
When you can define words as violence, then killing an alt-right paraplegic troll who can't feed himself much less raise a hand to defend himself, can be argued as "self-defense". The ubiquitous "punch a nazi" type of self defense, as it were.
Manipulating search results in China to favor one political point of view cannot be rationally distinguished from manipulating search results in the United States to favor one political point of view.
They don’t become hypocrites because they disagree with someone who believes as a matter of principle that non-white people should be harrassed.
If they believe that non-white people should be protected from harassment, but simultaneously believe that white people should *NOT* be protected from harassment, or are somehow incapable of being harassed because of their skin color, then they *are* hypocrites.
Here's the trick with free speech - it's either a principle you have to accept fully, or you really don't accept it at all. You can't have "sorta" free speech, in the same way you can't be "sorta" pregnant - you either are, or you aren't.
They do realize that google regularly censors results in America, right?
Any urgent moral or ethical issues with say, blacklisting Alex Jones? Down ranking alt-right sites? Artificially manipulating auto-completes to prefer one political party?
I'm not sure if these people realize that the "secret" work isn't just in China.
I'm not sure if that follows. The "normal" fully *expected* that the computer was going to take O(n^2) time. That seemed like a logical outcome for them - it wasn't a surprising result, and even seeing the exact, step-by-step path the program was taking would not have inspired them to imagine that there was a better algorithm for solving the problem.
I think this is exacerbated even more by the grand performance expansion of computer hardware over the past 30 years - you can get away with shitty algorithms simply by throwing more iron at it, and some people have no idea how to even *recognize* optimization potential, much less understand it's possible.
The people who understood big-O notation, and applied it regularly, never really talked about it much. The only context I would ever bring it up in would be about "what baseline skills are required to become a decent programmer", not "let's have a long code review and critique every loop for big-O optimizations".
Having a library of tools available, and knowing that you should use them are two separate things.
The "normal" guy didn't realize that searching for duplicates without sorting first was O(n^2). Giving him a sort function in a library isn't going to help him if he doesn't even realize there is a need to sort.
The "normal" would have seen exactly what they thought they would see - they had written a simple, reliable, completely unscalable algorithm.
They had no idea that it could be improved in scale. They had no idea that their solution, while *valid*, was not *optimal*. Simply showing them "hey, you're iterating through 1 million rows 1 million times" wouldn't have even registered with them as a problem - it was simply the reality of solving their issue logically and completely.
Now, how much training does it take to get someone to truly understand quicksort, bubblesort, or bucketsort? How much math do they have to understand to not only grok the *problem* with O(n^2), but also viable alternatives to their naive thought process? How much experience do they need to go through before they can apply this to arbitrary business requirements?
There's a book "Drawing on the Right Side of the Brain", which makes a great case that the art of drawing is not a mechanical one of hand dexterity, but a mental one of being able to see properly. It seems the same thing might apply here.
I guess I'd be really interested in what the minimal training would be to understand "big O" notation, implementations, and applications to reality. Maybe some people can get it in weeks, others, months or years.
Unless you can teach the average schmoe the difference between O(n^2) and O(nlogn), you've got no hope of bringing "programming" to the masses. I sped up a visualbasic "program" written by a "normal", by fixing their check for duplicate records by sorting *first*, then looking for dups, instead of going to each row, and checking every other row in the database for dups. Went from running 48 hours to less than 2.
Put another way, there are *lots* of ways to program poorly. Democratizing even the most basic tools won't prevent that.
Looks like Facebook needs to learn a little bit from China, who has done a bang up job of filtering offensive content in their country. I hear North Korea has a pretty good handle on locking down "problematic" content from the hoi polloi.
Those decrying the free speech rights of russian trolls might want to think about babies and bathwater for a while.
The cost of, for example, flood barriers are easy to find. The usage statistics for said flood barriers are also easy to find. This is just one very simple, not hard to think of, example that refutes your original statement.
Okay, so, give me the data from 1918 and 2018. Prove your point with data that you claim is easy to find.
I'll argue that you have neither of these:
1) a model that can correlate global average temperature to cost or usage of flood barriers by hindcasting the past
2) a model that can correlate global average temperature to cost or usage of flood barriers in the future
All of which is beside the point: Money is not the only thing that matters
Well, I'll take that a step further - money spent on flood barriers is not the only thing that matters. The use of natural petroleum and the Haber process to increase the carrying capacity of humans on the earth through improved agriculture also matters. The benefit of additional humans on the planet matters. Average quality of life of these humans matter. All of which has dramatically improved during a century of global average temperature increase.
Or are you a sociopath who believes we should be restricting human procreation, and need to return to the status of 1918?
Slashdot Mirror
...The Best President Of All Time.
Literally.
They don't just "sound" reasonable, they *are* being reasonable.
We have never had any economic models with any level of accuracy, ever, in the history of mankind. Asserting that you know "how damaging" *anything* could be to the economy, 80 years ahead of time, is wishful thinking at best.
It might not be satisfying to live with uncertainty, but that *is* the reasonable position - we don't know if increased CO2 emissions and any temperature increase due to that is going to be a net positive, or a net negative. Anyone who tells you otherwise has made a leap of faith.
Well, affirmative action privilege is great, right up until it hits the wall of meritocracy and skill :)
Now, if privilege == skill, well, that's a great position to be in. Skills don't care about your feelings :)
Not a troll, but yeah, this is going to be seen that way.
Once you've corrupted academia, and given degrees to people based on the color of their skin, or their SJW credentials, can you use degrees as a reasonable proxy for skills anymore?
Sounds like the meritocracy is going to work its way around attempts to thwart it.
I want a beef-based plant replacement.
My broccoli should be made out of cow.
Have they also asked for a non-censored search engine in the United States?
It seems to me that Google's activity in China is simply an expansion of their current US censorship, rather than anything new. Sure, the keywords might be "falun gong" instead of "conservative news", but it's silly to protest their activity in China without acknowledging their activity in the States.
I'm agreeing with you as hard as I can.
I think what you don't see is that excess consumption doesn't drive fat accumulation without insulin :)
Even further, I think what's interesting is that when you see a 500 pound fat man eating pizza like he's starving, it's because his muscles *are* starving, and his fat cells, under the influence of insulin, are stealing all the energy from his muscles.
Now, is it "excess consumption" if your muscles are starving? If your energy is being partitioned poorly into fat cells, it's actually a *survival* skill to eat more to provide your muscles energy :)
Yeah, you completely missed the Kreb Cycle.
Try applying basic physics to this problem:
Take an aluminum rod, 2 feet in length. Place one end of the rod in a pot of warm water. Measure the time it takes for the other end of the rod to increase in temperature and reach equilibrium.
Now, do the same experiment with a human. Place one hand in a pot of warm water. Measure the time it takes for the other hand to increase in temperature and reach equilibrium.
Now, of course, basic physics still works - but what you're really interested in isn't "calories in/calories out" as measured by "food in the mouth, physical activity of the muscles". You could put a metal ball with 10,000 calories of energy in your body, and not be able to metabolize any of it. So the *quality* of the calorie counts too.
In the case of insulin, which drives fat accumulation, it is driven by blood sugar levels, which is driven by carbohydrate intake (particularly refined carbohydrates). If you don't include that in your calculation, you're using a simple formula for a complex scenario.
Are you saying that it is impossible to program algorithms against SEO abuse without engaging in political viewpoint discrimination?
You keep conflating things - I'm not sure if you're engaging here in an attempt to understand a point of view contrary to your own, or just get a quick endorphin rush from arguing with someone on the internet :)
So, therein lies the rub I suppose - how do you propose to create a search engine that reads people's minds? Someone with the same search terms might be looking for something completely different than you are.
Now, maybe it's as simple as designing pre-calculated "safe-search"-esque modes where someone can "filter-out-right" or "filter-out-left" or "filter-out-sarcasm" or "filter-out-humor" modes. But assuming you can build a single mode that will in fact, satisfy the various intents of people who are resistant to mind reading...I'm not sure if that's helpful.
Why should you trust anyone to decide what is useful, and what is useless to you? Why should anyone trust you to make that decision for others?
Funny, google puts a planned parenthood link on the top of my results:
https://www.google.com/search?...
In fact, on that first page, not a single anti-abortion link comes up.
Now, before you freak out, please, understand this - I'm firmly pro-choice. I buy the rapist Bill Clinton's formulation of "safe, legal, and rare", and I'm not by any stretch of the imagination a pro-life protester. But this kind of censorship that Google does, in the United States, against political points of view, is brutally obvious.
I mean, brutally obvious.
Now, it may very well be that your search, done in the UK, is less censored that US results...but whatever your opinion on the matter, we should be incredibly reluctant to give powerful gatekeepers the ability to drive the conversation through editorial bias eliminating points of view from the material being presented. Any bit of joy you have over someone like Alex Jones being downranked is going to come back and bite you in the ass when a different political bias gains power, and works against your interests.
Well, I would make whatever algorithms being used transparent - obviously trivial stuff like keyword abuse recognition, and other metadata analysis that would indicate SEO abuse - however, I wouldn't allow for point of view discrimination. For those people worried that their view point was being discriminated against, they'd have the algorithm to look at, to double check that it isn't simply taking some blacklist of conservative celebrities, and shadow banning them.
The problem here is that the Overton window keeps getting thinner and thinner, and pushed further and further to the left. This isn't healthy or responsible.
So, you don't have a problem with returning an Alex Jones rant, when you search for "gay amphibians"? Or returning a wattsupwiththat.com article when searching for "global warming"? Or autocompleting "hillary clinton is...married to a rapist" if that is in fact the most common completion?
My fear here is that we've gone beyond "someone searches for amphibians and gets Alex Jones", and gone into "someone searches for abortion and only gets pro-choice views". Deciding what is "useful" for someone a priori seems like something we should be much more careful about.
Okay, thank you for responding to two of the lines.
Can I assume you agree with the third one?
"How about giving users the option to "opt-in" to a politically correct blacklist/whitelist, and allowing the default to give the raw results (sort of like "safe-search")?"
I'm not sure if anything would count as censorship by your definition, other than "well, censorship is when something I personally like is restricted".
Is there any content that you find distasteful, offensive, misguided, or otherwise incorrect that you *wouldn't* censor?
I'll start with my example - I think the bible is distasteful, offensive, misguided, and generally incorrect. I do not think any of it should be censored from search results. Can you come up with anything like that, so we can understand what you might *not* censor?
https://www.google.com/search?...
So...it's not censoring when there are exactly zero shopping results for confederate flags?
Really?
https://www.google.com/search?...
Or glock?
(ruger actually shows up with some holsters, but no guns)
Are you ready to change your point of view, now that you've been confronted with data that contradicts your initial opinion?
How about just having a content neutral algorithm that does a text search without any manual tweaking of the results in order to get a pre-ordained result?
How about giving users the option to "opt-in" to a politically correct blacklist/whitelist, and allowing the default to give the raw results (sort of like "safe-search")?
There are lots of ways of providing a search engine without being censorious.
This seems to be the exact position of China's censors.
Here's their constitution:
http://en.people.cn/constituti...
"Article 35. Citizens of the People's Republic of China enjoy freedom of speech, of the press, of assembly, of association, of procession and of demonstration."
You're falling down a deep semantic hole there :)
When you can define words as violence, then killing an alt-right paraplegic troll who can't feed himself much less raise a hand to defend himself, can be argued as "self-defense". The ubiquitous "punch a nazi" type of self defense, as it were.
Manipulating search results in China to favor one political point of view cannot be rationally distinguished from manipulating search results in the United States to favor one political point of view.
If they believe that non-white people should be protected from harassment, but simultaneously believe that white people should *NOT* be protected from harassment, or are somehow incapable of being harassed because of their skin color, then they *are* hypocrites.
Here's the trick with free speech - it's either a principle you have to accept fully, or you really don't accept it at all. You can't have "sorta" free speech, in the same way you can't be "sorta" pregnant - you either are, or you aren't.
If I was a Chinese censor, that is exactly the argument I would make against Falun Gong, or any number of dissidents.
They do realize that google regularly censors results in America, right?
Any urgent moral or ethical issues with say, blacklisting Alex Jones? Down ranking alt-right sites? Artificially manipulating auto-completes to prefer one political party?
I'm not sure if these people realize that the "secret" work isn't just in China.
They said this 20 years ago.
http://www.cnn.com/TECH/scienc...
I'll bet they'll say the same thing 20 years from now, no matter what happens with global average temperature :)
I'm not sure if that follows. The "normal" fully *expected* that the computer was going to take O(n^2) time. That seemed like a logical outcome for them - it wasn't a surprising result, and even seeing the exact, step-by-step path the program was taking would not have inspired them to imagine that there was a better algorithm for solving the problem.
I think this is exacerbated even more by the grand performance expansion of computer hardware over the past 30 years - you can get away with shitty algorithms simply by throwing more iron at it, and some people have no idea how to even *recognize* optimization potential, much less understand it's possible.
I am agreeing with you as hard as I can.
The people who understood big-O notation, and applied it regularly, never really talked about it much. The only context I would ever bring it up in would be about "what baseline skills are required to become a decent programmer", not "let's have a long code review and critique every loop for big-O optimizations".
Having a library of tools available, and knowing that you should use them are two separate things.
The "normal" guy didn't realize that searching for duplicates without sorting first was O(n^2). Giving him a sort function in a library isn't going to help him if he doesn't even realize there is a need to sort.
The "normal" would have seen exactly what they thought they would see - they had written a simple, reliable, completely unscalable algorithm.
They had no idea that it could be improved in scale. They had no idea that their solution, while *valid*, was not *optimal*. Simply showing them "hey, you're iterating through 1 million rows 1 million times" wouldn't have even registered with them as a problem - it was simply the reality of solving their issue logically and completely.
Now, how much training does it take to get someone to truly understand quicksort, bubblesort, or bucketsort? How much math do they have to understand to not only grok the *problem* with O(n^2), but also viable alternatives to their naive thought process? How much experience do they need to go through before they can apply this to arbitrary business requirements?
There's a book "Drawing on the Right Side of the Brain", which makes a great case that the art of drawing is not a mechanical one of hand dexterity, but a mental one of being able to see properly. It seems the same thing might apply here.
I guess I'd be really interested in what the minimal training would be to understand "big O" notation, implementations, and applications to reality. Maybe some people can get it in weeks, others, months or years.
Unless you can teach the average schmoe the difference between O(n^2) and O(nlogn), you've got no hope of bringing "programming" to the masses. I sped up a visualbasic "program" written by a "normal", by fixing their check for duplicate records by sorting *first*, then looking for dups, instead of going to each row, and checking every other row in the database for dups. Went from running 48 hours to less than 2.
Put another way, there are *lots* of ways to program poorly. Democratizing even the most basic tools won't prevent that.
Looks like Facebook needs to learn a little bit from China, who has done a bang up job of filtering offensive content in their country. I hear North Korea has a pretty good handle on locking down "problematic" content from the hoi polloi.
Those decrying the free speech rights of russian trolls might want to think about babies and bathwater for a while.
Okay, so, give me the data from 1918 and 2018. Prove your point with data that you claim is easy to find.
I'll argue that you have neither of these:
1) a model that can correlate global average temperature to cost or usage of flood barriers by hindcasting the past
2) a model that can correlate global average temperature to cost or usage of flood barriers in the future
Well, I'll take that a step further - money spent on flood barriers is not the only thing that matters. The use of natural petroleum and the Haber process to increase the carrying capacity of humans on the earth through improved agriculture also matters. The benefit of additional humans on the planet matters. Average quality of life of these humans matter. All of which has dramatically improved during a century of global average temperature increase.
Or are you a sociopath who believes we should be restricting human procreation, and need to return to the status of 1918?