Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Escape the echo chamber

aeon.co

Post reply on HN
Print view View original

31–40 of 69 posts

Re: Escape the echo chamber

#31

Post by jamesblonde »

I believe that there are even cases of echo chambers for systems software out there. For example, the database community is pretty tight with a center of gravity around Stonebraker (Postgres, Vertica for columnar stores, voltdb for NewSQL dbs). Well, there's a DB out there for OLTP workloads called NDB (MySQL Cluster) and not only is it the fastest in the world (200m transactions/sec on commodity hardware), but it is also GPL-v2 licensed. But, it's completely unknown in the valley. And when evidence is presented about it it, the evidence is ignored and conflated "in echo chambers, other voices are actively undermined"

Reference: http://mikaelronstrom.blogspot.se/2015/03/200m-reads-per-sec...

Re: Escape the echo chamber

#32

Post by majewsky »

drjesusphd wrote: post #30
lopmotr wrote: post #27

Earlier quoted context omitted.

Absolutely. Even worse though, is the fact that ivy league colleges aren't growing much bigger. If they're so good at educating people, and have so much donation money, why not expand? I think they intentionally limit their size to maintain their elite status, which shows they're more of a zero-sum selection filter than simply good at educating.

How is MIT's Open Courseware and the many programs like it not exactly such an expansion in line with the mission to educate? The idea that success == expansion is a capitalist (or even more narrowly, Wall Street) ideal and Universities should not be run like businesses.

People are becoming more and more educated over the world, not just because of population increase, but also because the world in general is becoming more prosperous. [1] It therefore makes sense for universities (like all other levels of education) to expand to meet the increased demand. More education can be argued for from a capitalist perspective, a communist perspective, a humanist perspective, etc.

[1] e.g. https://ourworldindata.org/tertiary-education

Re: Escape the echo chamber

#33

Post by SiVal »

stillkicking wrote: post #20
3131s wrote: post #10

Earlier quoted context omitted.

Tell them that it is completely impossible to set up rigorous controls in a study of ethnicity and intelligence anyway. The bigger question is why do some people care so much and why are they so focused on ethnicity among many other data points that could be predictive of intelligence. I typically ask these people if some connection between race and intelligence actually existed, is it your intention to change how so…

Why care? That seems pretty obvious: because ethnicity is already used to justify policies of affirmative action, used as a measure of prosperity and equality. The common reasoning is that until the conditional probabilities by demographic are flat, there must be systemic oppression which must be rectified by applying concerted pressure in the other direction. This is an a priori assumption that there are no relevant…

Yes, exactly. All selection policies (hiring, admissions, contracting, etc.) that say that choosing the best needs to result in proportional representation or it is cheating (not really choosing the best but choosing the "privileged") are based on the ridiculous assumption that all groups have the same mean "goodness" despite variation in "goodness" among individuals.

Many of these policies become laws with significant punishment specified for violators.

In fact, if there is individual variation, and the groups are clustered based on something more significant than random partition, such as genetic similarity, the groups themselves are frequently going to exhibit differing statistical tendencies in any heritable characteristic because of that genetic similarity.

Whether these statistical tendencies differ or not is an empirical question, not a question of political preferences.

For law, and coercive power of punishment, to be based on reality rather than political favoritism, the question of whether the group differences are real or not matters. If the group differences are real, punishing people for making choices (again, hiring, admissions, etc.) based on reality instead of on political theory is the real injustice.

So the actual, measured average IQ differences between groups matters if you are going to use any assumption about it as a basis for policy decisions.

Re: Escape the echo chamber

#34

Post by jonathanstrange »

3131s wrote: post #10
jackfraser wrote: post #7

Earlier quoted context omitted.

> And they would come back with 150 more recent, more well researched studies and explain to me how statistics works Has anyone actually put this data together in a presentable way? I've only seen things that say the opposite and it would be good to actually have a read of something well-formed that sets the modern progressive case for dealing with this research, beyond simply ignoring it altogether.

Tell them that it is completely impossible to set up rigorous controls in a study of ethnicity and intelligence anyway. The bigger question is why do some people care so much and why are they so focused on ethnicity among many other data points that could be predictive of intelligence. I typically ask these people if some connection between race and intelligence actually existed, is it your intention to change how so…

It's an empirical matter, but my hunch is that most of the people who focus on correlations between intelligence and ethnic groups or phenotypical perceived "race" are racists. Many people aren't aware that you can be a racist without knowing it. (In a broader sense we're all racists, as practically everybody overgeneralizes based on stereotypes.)

It's important to stress that the differences between members of the perceived groups ("ethnicities", "country of origin", "mother tongue", "a certain look") are much larger within that group than any measurable differences between averages of those groups. That is not to say that some research in these areas cannot make sense for e.g. policy making. For example, people who have suffered severe malnutrition in their childhood can have lower IQ on average, and the same is the case for many other environmental and social factors, of course.

The reason why people like to focus on intelligence is that unlike other human characteristics like strength, intelligence is almost universally used as a kind of honorific. If you tell someone that he or she is intelligent, this will make them proud. If you tell someone he or she is not very intelligent, this will be taken as a grave insult. Additionally, everybody overestimates his own intelligence.

All of this is kind of ironic, because intelligence is highly overrated. Intelligence doesn't count much, it's what you make of it. Moreover, whether you like it or not, society does not have a need for more highly intelligent people. There are not enough jobs for them.

Last but not least, people who think of themselves as being very intelligent usually haven't tried to solve hard enough problems.

Re: Escape the echo chamber

#36

Post by 4bpp »

3131s wrote: post #10
jackfraser wrote: post #7

Earlier quoted context omitted.

> And they would come back with 150 more recent, more well researched studies and explain to me how statistics works Has anyone actually put this data together in a presentable way? I've only seen things that say the opposite and it would be good to actually have a read of something well-formed that sets the modern progressive case for dealing with this research, beyond simply ignoring it altogether.

Tell them that it is completely impossible to set up rigorous controls in a study of ethnicity and intelligence anyway. The bigger question is why do some people care so much and why are they so focused on ethnicity among many other data points that could be predictive of intelligence. I typically ask these people if some connection between race and intelligence actually existed, is it your intention to change how so…

At this point, I get the impression that for most (by numbers) of those who believe in a race-$trait correlation, the important thing is not to persuade those who are indifferent but to prove those who strongly assert that there is no such correlation wrong. More often than not, for each desirable trait there is cold hard real-world data that suggests strong correlations between race and particular desirable outcomes that appear related to the trait in question. (Say, in the case of IQ, lifetime income, job prestige etc.) If you believe that this difference in outcomes can be traced to a genetic difference in causative traits, to a first approximation the data does not suggest doing anything (with future technology, maybe, it would suggest a large-scale programme of gene therapy). If you believe that there are no such genetic differences, on the other hand, the data implies that there is a great and particularly morally repugnant defect to society, and justifies extraordinary, often outright punitive (e.g. racial quotas) corrective measures.

Currently, as far as policymaking in western societies is concerned, the faction that follows the "no genetic differences => (data => extreme measures are warranted)" reasoning is clearly in charge, if you consider e.g. the recent altercations about racial quotas in school discipline (article for: https://www.nytimes.com/2018/03/18/us/politics/school-discip... / against: https://www.city-journal.org/html/no-thug-left-behind-14951....). If you read the "against" article, that should make it pretty clear why those who believe in the existence of group differences consider the answer important, if "it's important to not do an unwarranted extreme thing" is not already implied by the term "extreme". In the light of this, it does not seem fair to imply that the only significant societal changes a belief in genetic group differences would imply are ones that are so terrible that anyone would be embarrassed to admit advocating them to you or even themselves.

Re: Escape the echo chamber

#37

Post by primitur »

arkh wrote: post #21
3131s wrote: post #10

Earlier quoted context omitted.

Tell them that it is completely impossible to set up rigorous controls in a study of ethnicity and intelligence anyway. The bigger question is why do some people care so much and why are they so focused on ethnicity among many other data points that could be predictive of intelligence. I typically ask these people if some connection between race and intelligence actually existed, is it your intention to change how so…

If IQ is closely related to genetics and IQ is important for your society (two bold assumptions) you could argue for some eugenics and advantage high IQ people for reproduction. But as you say it could go really bad really fast. Especially if some of those high IQ genes are linked to some illness. For your question about why some people are focused on ethnicity? As usual: people's appearance are easier to discriminat…

>argue for some eugenics and advantage high IQ people for reproduction

See .. I find this line of thinking heinous. Why do you not instantly trip up when you type this?

Re: Escape the echo chamber

#38

Post by 4bpp »

jonathanstrange wrote: post #34
3131s wrote: post #10

Earlier quoted context omitted.

Tell them that it is completely impossible to set up rigorous controls in a study of ethnicity and intelligence anyway. The bigger question is why do some people care so much and why are they so focused on ethnicity among many other data points that could be predictive of intelligence. I typically ask these people if some connection between race and intelligence actually existed, is it your intention to change how so…

It's an empirical matter, but my hunch is that most of the people who focus on correlations between intelligence and ethnic groups or phenotypical perceived "race" are racists. Many people aren't aware that you can be a racist without knowing it. (In a broader sense we're all racists, as practically everybody overgeneralizes based on stereotypes.) It's important to stress that the differences between members of the p…

There is a lot I am tempted to call out in this post, but just for a particularly glaring example,

> It's important to stress that the differences between members of the perceived groups ("ethnicities", "country of origin", "mother tongue", "a certain look") are much larger within that group than any measurable differences between averages of those groups.

Surely the differences in income, the binary variable of being-in-STEM, recruiter favourability shown towards CVs etc. between members of the perceived groups ("skin colour", "gender", "mother tongue", ...) are much larger within that group than any measurable differences between averages of those groups too. Does that mean you are arguing that the gender pay gap, representation in STEM fields and unfair hiring practices are all irrelevant?

(For what it's worth, there's only a few examples where differences between groups actually dwarf differences within groups, all of which seem very famous, e.g. grip strength vs. gender.)

Re: Escape the echo chamber

#39

Post by primitur »

drjesusphd wrote: post #30
lopmotr wrote: post #27

Earlier quoted context omitted.

Absolutely. Even worse though, is the fact that ivy league colleges aren't growing much bigger. If they're so good at educating people, and have so much donation money, why not expand? I think they intentionally limit their size to maintain their elite status, which shows they're more of a zero-sum selection filter than simply good at educating.

How is MIT's Open Courseware and the many programs like it not exactly such an expansion in line with the mission to educate? The idea that success == expansion is a capitalist (or even more narrowly, Wall Street) ideal and Universities should not be run like businesses.

>success == expansion is a capitalist

Well, is it capitalist really, or is it something that capitalists are guilty of propagating as a means of control?

I mean, it could be said that 'success == expansion' is more of a natural law, which can be applied to any 'ism and used, either to contract agency, or .. grow it.

Re: Escape the echo chamber

#40

Post by Vinnl »

ScottBurson wrote: post #6

Look at the many stories of people leaving cults and echo chambers. Take, for example, the story of Derek Black in Florida – raised by a neo-Nazi father, and groomed from childhood to be a neo-Nazi leader. Black left the movement by, basically, performing a social reboot. He completely abandoned everything he’d believed in, and spent years building a new belief system from scratch. He immersed himself broadly and ope…

I think in terms of the context of the article, it is true though:

> the member should temporarily suspend all her beliefs – in particular whom and what she trusts – and start over again from scratch

That most important one -whom and what to trust- was "rebooted", which allowed Black to be open to those 150 more recent, more well-researched studies, rather than dismissing them as the echo chamber would tell him to do.

(I'm mentioning this because I noticed by myself that I almost dismissed the article due to first encountering this comment, and ended up enjoying it - which is an interesting observation in the context of this article by itself.)

Post reply on HN