/sci/ - Science & Math

Academia is just a bunch of pompous no talent midwit blowhards. If they were really as smart as they say they are then they would be able to produce something other than vanity publications that nobody reads

>>15771903
Who provided the data for this graph?

>>15773351
Nature

>>15771903
peer review doesnt work

>>15771903
>the crisis only applies to soft ""sciences"" like psychology, sociology and medicine
Nothing burger. I sleep.

>>15773683
"peer review" is just another way of saying "collusion"
"reproducibility crisis" is just another way of saying "falsification epidemic"

>>15771903
I actually just thought about this. Isn't this only an issue in fields like psychology?

>>15771903
Hot take: a reproducibility crisis should expected; it's proof the scientific method is working. peer review contesting the findings of past research is what whittles away the excesses of conjecture to produce stronger theories.

>>15775210
>Peer review
>Seems good
>Attempt to reproduce
>It was bullshit all along
The reason we have a reproducibility crisis in the first place is because peer review is insufficient and doesn't catch falsehood. In at least some fields, it exacerbates it.

>>15772945
Well-stated.

>>15773683
The popularity of peer review, which started in the late 1960s when the boomers wanted to review each other instead listening to their more experienced and knowledgeable elders, is what inevitably lead to the "reproducibility crisis"
"Reproducibility crisis" itself is another boomerism, its a PC term they invented to make outright fraud by scientists sound like it was just a big accident by honest academics when its really intentional deceit

>>15775400
Checked.
What about the expansion of science though? As time goes on fields get bigger and specificity increases. It seems like there needs to be some kind of check in place.

>>15775400
Whats funny is that the people who all claim that they hate boomers are the same people who will say that nothing counts if its not peer reviewed

>>15771903
>I used to look up you guys in academia, I used to think that you are better and smarter than me
there was never hope for you

>>15774780
no

>>15772945
fpbp
the actually intelligent are in industry making money hand over first

>>15771903
>Maybe I am the smart one. I earn more than you, after all.
This was the line that killed your post. Money earned does not equal intelligence.

>>15777233
> Money earned does not equal intelligence.
Thats because money earned > intelligence

>>15771903
>I used to look up you guys in academia
I'm not in academia and I take offense at you suggesting I am.
Apologize

>>15771903
>I used to look up you guys in academia
naïve

>>15774749
Does it? Or are those studies just easier to (attempt to) replicate?

>>15777527
Im actualy in a macadamia

>>15774749
>the crisis only applies to soft ""sciences"" like psychology, sociology and medicine
false

>>15779395
>macadamacadamia
the school of theoretical paving techniques

>>15775240
>In at least some fields, it exacerbates
>some
all

One of the common characteristics of midwits is that they're less capable of and less likely to concern themselves with the long term consequences of their own actions. Given that peer review is necessarily a midwit activity (because it involves votes from a group of people and the votes are not weighted by IQ, so the consensus if going to be decided by the group with the most common levels of intelligence rather than the highest), the reviewers are less likely to be concerned by the potential future results of dishonest behavior.
https://www.youtube.com/watch?v=d2Tm3Yx4HWI

>>15775240
>peer review is insufficient
There have been cases in which papers have been published even though reviewers clearly said something was odd and data was missing.
The problem might not be peer review... It's corruption.

>>15782971
The problem is in the authors, the reviewers, *and/or* the publishers. Requiring all three to be competent and truthful, often against their own self interest, means the system is broken and that's the absolute most charitable term for it.

However this is not technically a crisis. There is always the option to ignore it and go gently into a new dark age of ignorace and lies, exalting a new gospel of astroturfed fabrications.

>>15782984
the reason they can presume that being untruthful is in their self interest is because they think that being untruthful isn't a mortal sin. the scientific method was developed by Christians and for Christians, other groups cannot execute the method, science is a subset of Christianity

>>15771903
I have a few published papers. When submitting to journals, most reviewer comments were full of typos and gave the impression that they didn't really read the paper fully. There were basic misinterpretations of the intent of experiments, requests for new experiments that at best would be some supporting info figure that added very little to the work. I never got good feedback, it was either generically positive, or generically dismissive.
I wrote a review at the request of my PI and the comments on that were mostly
>paper is good but author failed to cite important works X, Y, Z
Then you check X Y Z and it's all from the same author and then I'm just like oh so the reviewer wants citations on their work, add the citations and the resubmission is accepted.
Those were the core of the research I was proud of in the first few years, after that my prof joined an EDI board and was largely absent and my projects started to fall apart because the lab had no focus. I did my best to drive my projects independently but we would waste hours and hours at lab meetings reviewing not the results of experiments but what my prof thought the story of the research was supposed to be. I submitted two more bogus garbage papers in those times and they had the same kind of comments as my good research. The whole experience just left me feeling like it doesn't really matter what you put out, you'll get a few generic comments back and you make a minimal effort to address them and then it's accepted. These aren't shit journals either, this was impact factor 20-30 materials chem. There are exceptions though, like I knew a guy who got a good nature paper that resulted in several patents and he's doing very well now.

>>15774780
No, biology is fucked too

>>15777233
income correlates with IQ

>>15771903
what happens as my friend explained it.
scientist or people in academia,
only care about being right, they are not interested in the truth.

neither are they interested in disproving their own
hypothesis.suffering from confirmation bias,when publishing research paper.


this was evidently clear when entering higher education as students where encouraged by the professors, to throw away and remove all results
that disapproved there hypothesis

progress for progress sake

The problem is we have a good two centuries of liberal enlightenment brainwashing telling us science is the only credibility and so we try to use it on things that it simply cannot process properly. Instead of accepting this and using some other kind of investigation people try to ram it in to science diluting the accuracy.

Science should never have been seen as a way to discover all things. Its a paranoid reaction to the fear of uncertainty. Some things can be proven, we can know them. This is what the task is. Not producing good arguments or speculations or making yourself look attractive to investors. Its finding the few things that we dont have to constantly worry we are deluding ourselves about.

You should just not consider something true if its not had 2 or 3 replication attempts confirming it. minimum

>>15774780
its a fucking huge problem in psychology?

many people in med school, and doctors
geniuly believe the "chemical imbalance" theory

funny thing it have never actually been proven
in the last 40-50 year. Its called an urban legend.
its actually just a hypothetical explanation.

>computer science
So you're working in some math/engineering adjacent field.
The replication crisis doesn't concern these as much, especially not mathematics. But if lumping in actual hard math based fields alongside bullshit like sociology makes you feel like a big man, well.... I can't stop you, but you should probably feel pretty silly.

>>15784688
How do you cope with MBA's making twice your salary?

>>15784720
I don't understand why the question is relevant?

>>15784723
It isn't an individual MBA. On average, MBA's will earn significantly more than a PhD in mathematics or Computer Science. Most MBA's have no interests in academic pursuits or in utilizing analytical skills to make their way through life or occupation. Most will make their way through entirely based on socialization.
If you are okay with accepting that MBA's have higher IQ's than math, CS, and Physics PhD's on average, than I guess I have nothing else to say since you already knew this. If not, then there is a conflict in your world view. That is all I am trying to say.

>>15784688
yeah sure. But its not the best predication of
income.

>>15784731
The correlation is stronger within fields. between fields there are more compounding variables.

>>15771903
>SWE
borderline retarded, but capable enough to palms head against keyboard and not choke on mouse

>>15784720
how do they cope with professional athletes making 100x the standard soience salary?

>>15784770
professional athletes must have 100x higher IQ than them

>>15771903
>Perhaps you are just publishing meaningless papers because this is the metric that you are measured by.
Sadly it's this. Actual, genuine research takes years, maybe decades, but grants are given based on the number of papers my physics department publishes, no matter if they're total bogus.
So every 6 months a few post docs have to write some shit that makes no sense and publish it just to keep the number of papers high enough to get funding for their actual work.

>>15785087
>to get funding for their actual work…
…which will never amount to anything anyway, what a total waste.

I'm an electronics engineer and I designed and implemented the instrument for a scientific mission recently

It's a bit interesting that the scientists will get all of the credit even though they would have had nothing if they did not have an engineer to do the actual work. It is career defining for them, even though the actual job that sets this project apart from other similar projects was getting data. Any random student can plot data and hypothesize

I'm not bitter or jealous, I just find the dynamic interesting. I know a scientist at NASA that does not even give his (very talented) engineers a co-authorship in his papers that only happened due to their design

>>15786267
but you know who takes the blame if instrument doesn't work out

>>15774780
Do you have any idea how bad it is in chemistry?
Imagine you spend 12 months devising a new synthetic methodology, but it relies on this super special catalyst you can either blow a half a year on making in useable quantities.
You publish a paper with a scope of 50 different examples of thst reaction.

How is anyone going to know if you 1. actually did all of those and 2. aren't overreportig yields?
Because in truth, nobody does all of them and everyone makes their yields look more impressive than they are.

Chemistry is particularly bad, because it is expensive.

Have Hillman's concerns about biology been addressed yet?
https://www.youtube.com/watch?v=GV1niiqmjU0

>>15772945
FPBP

>>15787189
Its the same everywhere, researchers compete to see who can demand the most welfarebux gibes for their research so they design the fanciest, most expensive experiments they can and the end result of that is that nobody ever bothers to reproduce the experiments. because everyone knows their work will never be tested or reproduced, every experiment is deemed a success by the people who carry it out and they all get the results they were hoping for before the experiment took place.
everyone always wins, everyone gets a trophy publication in the academic vanity press, everyone has a stellar reputation and nobody is ever wrong.

The problem is that 90% of all research is either funded by a marketing company for a product or it has an inherent political bias. You can reproduce any of these studies because they didn't follow the scientific method from the beginning. The conclusion is written before the study is even started.

>>15788857
>The conclusion is written before the study is even started.
Thats what happens when you let atheists call themselves scientists. Atheists are all dishonest people, so they will all use every dishonest form of abuse their tiny little minds can think up

>>15788779
>everyone always wins, everyone gets a trophy publication in the academic vanity press, everyone has a stellar reputation and nobody is ever wrong.
and if someone from outside the soience priest class dares to question the soience dogma, thats heresy

>>15775210
>>15771903

Is the reproducibility problem for published peer reviewed research papers?

>>15773351

>>15772945
>they would be able to produce something other than vanity publications that nobody reads
This really hits me hard as a PhD student.
... because it's mostly true.

>>15791089
>asking for a source is considered a rude remark in a post-ironic stage of the internet era
0-0 L

>>15789943
>so they will all use every dishonest form of abuse their tiny little minds can think up
They learn a lot of their bad behaviors from the jewish media they consume. Thats why the media is always portraying noxious and toxic behavioral characteristics positively, in order to convince the viewers to behave that way.

>>15775210
retarded take. thats like saying having 50% of your population in prison is a good thing because youre getting rid of all the prisoners. youre not understanding the issue at hand.
when a study comes out that fails to reproduce the results of a previous study, you didnt make progress towards the truth. youre just back to square one. if half the research in a field cant be replicated, then theres no progress actually being made in that field. its a car spinning its tired in the mud.

>>15771903
Since there probably are not even five people in this thread who have ever taken part in an academic peer review process, I will enlighten you.
Writing a good review takes a lot of time. You have to read the paper, understand it, check formulas and figures, check the related work to make sure the contents of the paper are actually novel and also not obvious plagiarism. This takes several work days. Those are work days you cannot use on your own research - and your own research is the only thing that matters for your own academic success and career. Nobody cares how many good peer reviews a researcher wrote.
That is why most reviewers just skim a paper and look for obvious inconsistencies. Afterwards, they write a few meaningless lines about it. In total that maybe takes two hours and the reviewers have more time for their own research.

>>15790677
Yes.

>>15792915
>Since there probably are not even five people in this thread who have ever taken part in an academic peer review process, I will enlighten you

>>15794134
He's not wrong, though.

NPC drone thread

>>15771903
>I can't reprodooooce
this is what happens when chuds become scientists

>>15772945
fpbp
/thread
this woman made more for science than all of them combined.

>>15789943
>>15792712
>athiests
>jews
nah, casuistry starts from a conclusion and argues backwards from it. Leave your /pol/ shit at the door, organized religion is the problem, while the scientific method came from the idea that the truth is true no matter who says it, or how loudly, or by someone well-armed. This notion, radical at the time, allowed rigour to be a thing.

>>15794460
Literally who?

>>15771903
>a 'Nature' survey ...
>>15773351
>who provided the data for this graph
>>15791875
when something has its source as its first sentence but reading is hard and you can't quite make it that far, eh?

>>15794506
lurk more newfag

>>15794756
>image of random woman with hippie t-shirt
???

>>15794823
>random woman
>doesn't know who Alexandra Elbakyan is
lurk more newfag

>>15794506
Lurk more newfag

>>15794823
>random woman
retard, she created sci hub because computer scientists are the only scientists who actually do anything of real world value

>>15794951
If you do professional research, your institute usually has publisher subscriptions. Those aren't even that expensive. E.g., see here:
https://libraries.acm.org/subscriptions-access/academic/dl-pricing
$30k per year and every student of a university can legally access as many ACM digital library files as wanted.

>>15795609
In addition, here are the prices for companies:
https://libraries.acm.org/subscriptions-access/corporate/dl-pricing
The maximum price for companies with a size comparable to Microsoft or Apple is ~$130k per year.

There is just no need to illegally access research papers if you are a professional researcher.

>>15795629
>>15795629
>The maximum price for companies with a size comparable to Microsoft or Apple is ~$130k per year.
> t. Elsevier shill

Why the fuck would a university or company waste an entire student tuition worth when they can just pirate it?

>>15795671
Because it's a crime. Additionally, at least parts of the fees are used to finance the file servers and the peer review process.

>>15795609
>your institute
academia is pozzed, long live scihub

>>15795880
Well, try to get a paper published as independent researcher.

>>15795883
if high cost hardware is not needed (only reason to be in academia desu), most of real future discoveries will be made by independent researchers.

>>15795885
And how are those independent researchers supposed to feed their families while they do research?

>>15795891
do some other job in parallel or work part-time, idgaf

>>15795883
whats so great about getting a paper published in a journal thats full in """research""" that isn't replicable?

>>15797063
Somehow you have to present the results of your research so other people can build on it and humanity can advance as a whole. Publishing the results as a paper in a journal or on a conference are best for that due to their high visibility.
Of course, you could also publish a paper here on /sci/ or on your own private website, but then the visibility is basically zero.

>>15797336
if these "other people" are academia midwits it's as good as useless.

>>15795883
theres nothing meaningful or useful about getting a paper published unless you have something worthwhile to communicate. virtually nobody in the world of science has that and they don't seem to bother looking for it too hard either

Is this whole board just filled with people who are bitter because they weren't able to join a PhD program and become professional scientists?

>>15771903
>But hearing about the replication crisis and looking into research practice has made me doubt my beliefs. Maybe you academics aren't actually changing the world, at least not the majority of you. Perhaps you are just publishing meaningless papers because this is the metric that you are measured by.

actually i clicked on this board just for this reason
i was doing research on my own and found my uni's researchers on some website and i found some of my professors (whom i dont think are very smart) having hundreds of papers published and i kinda felt a bit of a retard

is it really impressive having that many papers published or do they just do it for clout ?

i really hate the academic system and i don't think 99% of these help anyone in the real world but im still curious

>>15799039
When PhD students write a paper, they usually have to name their professor as co-author. Even if the professor hasn't contributed anything. That's how they get so many publications.

>>15799049
oh that makes sense

>>15798715
Not at all. Some of us got a PhD, did a postdoc and then went to industry. I am happy with my choices.

>>15798715
>professional scientists?
nobody is bitter because they were prevented from earning a five figure income

>>15800167
Right, because the highschool students and college dropouts on /sci/ earn any more than minimum wage.

>>15800521
>I hate 4chan
>everyone on 4chan is total losers and idiots
why are you here?

>>15800569
the online activist interlopers from reddit and leftychan.net and shareblue can ever answer this question

>>15771903
I don't see how publishing papers should considered valuable or worthwhile even if the papers weren't all total lies

>>15795673
>the peer review process.
which has proved to be a failure and a toxic hindrance to scientific progress.
peer review is the equivalent of maoist struggle sessions. china under mao was a massive failure, china's gdp per capita was on par with subsaharan africa into the 1990s due to the toxic idiocy of the maoists

>>15795673
>Because it's a crime.
its because they're not spending their own money and have no motivation to be frugal. people who have public sector jobs dish of a fortune of other people's money to private sector leech organizations and then after a while the private sector leech organization is ready to hire them for a better paying job than what the public sector offers. the military works the same way, whoever spends the most government money while in the service ends up as executive at a big military contractor as a bonus

>>15792712

>>15802232
science is a lot like modern art. they've both degenerated from the time back in the 1800s when the artists or scientist was supposed to produce something admirable and noteworthy or be laughed out of the field to the present state where the artists or scientist just produces whatever falls out of their ass and you're supposed to respect it because its art or science, respectively.

>>15805155
Good observation. Art and science both died in the early 1900s and both died at the hands of the same group

>>15773683
not in the way you think it's meant to, but its not intended to work to improve the quality of publications or scientific work, its meant to be a form of gatekeeping to prevent nonconformist thought from being published.

>>15773683
we live in a new era of lysenkoism.

>>15771903
I understood picrel at an early age, which is why I avoided academia like the plague.
I wish midwits would catch up fast so we can end this tower of bullshit.

>>15807269
>I wish midwits would catch up fast so we can end this tower of bullshit.
they won't because the "I'm smart, my university degree proves it" argument pleases them emotionally and they are animal like, emotionally driven creatures.

>>15771903
>I used to look up you guys in academia, I used to think that you are better and smarter than me
You were under the spell of thousands of incidents of media exposure which told you that. Never once did you ask yourself "hey why does the media keep on repeating this same message over and over and over again?"
Repetition is the single most powerful brainwashing technique, thats why the message got repeated to you so often.

>>15795891
by being born in rich families

>>15798715
>grow up hearing about space, plasma research, aether, chemical reactions, bioengineering, robotics, etc.
>work hard to get into a good school
>very few people actually care about advancing the field; those that do are boxed into a niche by office politics
>turns out that everything you have dreamed of has already been created as early as 70 years ago
>it was all shut down by bureaucrats to protect their financial interests
>realize that the upper echelon of your sector is a bird cage and until they collapse you're just a fancy wagie overworking your brain
You can still pursue whatever catches your interest and free-lance until you find a comfy job, but it's disappointing that there's an actual glass ceiling on innovation that is superseded by the 'white man bad' corporate structure one.

>>15805155
at one point in time going back about 200 years, artists and scientists both had roughly equal levels of technical proficiency. those were the days when jacquard was inventing the automated loom that evolved to become the modern computer and morse, a painter, was creating telegraphy that evolved to become packet switched data transmission.
these days neither artists or scientists do anything of much use, neither group ever develops any real concrete technical skills either

>>15771903
>I used to look up you guys in academia, I used to think that you are better and smarter than me
Just look at how rare it is to be 2 or 3 standard deviations above average. Then look at how many people are in supposedly high IQ positions.

IT DOESN'T FUCKING ADD UP MANG, THEY CAN'T ALL BE SMART

>>15774749
they don't even make you read the full textbook anymore

>>15774749

>>15772945
fpbp

>>15771903
i don't see what the replication crisis has to do with anything, it's only really a problem in psychology and some related parts of medicine and economics. not anything anyone on /sci/ should be doing and certainly not something relevant to anything a would-be SWE would be researching

>>15784731
What other human attributes corelate to income better than IQ?

>>15784734
Are we going to pretend that all sociologists and historians could have been MD's if they really wanted to?

>>15811036
how much truth is in this post?
as a scientist but not physicist I'm shocked and worried now.

>>15784688
*JQ

>>15795673
I've done peer review for over a dozen papers now in a few different journals. No one is getting paid to do reviews of submitted journal papers. It's all volunteer "paying your dues" kind of stuff except for the (generally under 10) people who literally work for the journal and don't actually do reviews themselves.

>>15812058
If people started getting paid for peer reviews then they might start taking it seriously and that would prevent a lot of useful idiot campus liberals from getting their absurd lies published. its done the way its and kept a virtual rubber stamp with a veneer of authenticity for good reason.

>>15813130
>If people started getting paid for peer reviews then they might start taking it seriously
but wouldn't they be incentivized to do as many as possible potentially disregarding review quality?

>>15794506
my wife <3

>>15772945
>vanity publications that nobody reads
Nobody reads them because everyone knows they're all fake

>>15815385
Counterpoint: the presence of terminally dissatisfied feminist bitches and their negro, hispanic, homosexual and jewish cohorts is what make academic workplaces toxic to begin with.

>>15794936
lol

>>15771903
The problem is that everyone has to become published to feel accomplished. We should prioritize teaching as a goal over original research. The results are unhealthy and clearly muddying the waters of all research

>>15794460
i will try to guess with no google. Is that russian bitch who invented sci-hub?

>>15815385
if they're so smart how come they can't figure out a solution to their own problems?

>>15816014
imagine getting a sense of accomplishment from publishing something that even the peer reviewers barely bothered skimming.
if you publish something and nobody cares, thats a failure, not an accomplishment

>>15817752
Works for me!

>>15818247
why?

>>15818607
Every published paper makes my science boner grow larger.

>>15819006
HAHAHAHA HE USED A DICK EUPHEMISM OUT OF CONTEXT HAHAHAHA WHAT A CARD!!! HOW DID HE EVEN THING THAT ONE UP!?!?!
OMG SO BRILLIANT
IQ = 9000

>>15805155
Yes, pretentiousness and bluster has replaced skill and honesty. And its all due to the lack of religious belief and education. The scientific method was built on the foundation of Christianity, with that foundation gone, there is no real science, instead theres a bunch of charlatans shilling lies and snake oil

>>15812058
cuck

>>15771903
>I used to look up you guys in academia
why would anyone ever admire a bunch of twisted, mentally ill nerds? too much exposure to bill nye on TV? the cringey nerds are constantly lying about pretty much everything because they're so ashamed of themselves.

The average IQ of the posters in this thread must be around 80

>>15784714
>chemical imbalance
Meme to sell chemically lobotomizing meds that are addictive and increase symptoms when you try to get of them.
Then when withdrawl symptoms hit they don't say its withdrawl, its the "meme" balance again.

There can be psychological issues due to "imbalance" its called being poisoned with neurotoxins.
Such as:
>flouride
>mercury
>alcohol
>aluminium
>arsenic
>bhc
>antimony

Etc.
But they will always say:
>oh no its only trace amounts
>it does nothing
>trust me
>its because your brain is defect
>its genetic
>please take sexdrive killing, side effect riddled chemical lobotomization subscription for life.

>>15816008
The elusive autistic female

>>15822998
Why does this thread upset you so much?

>>15824355
Because the posters here are generalizing too much. Yes, there are a lot of bad publications, especially in social sciences. Yes, even STEM journals and conferences have codes of conducts nowadays, but unless you write racist slurs or similar stuff in a paper, it will not get rejected because of the code of conduct. Good science that helps humanity to progress is still possible in STEM fields.
tl;dr Saying that all published papers are bullshit is an oversimplifying generalization.

>>15824372
you're only saying that out of self interest, you don't care even slightly about "helping humanity", thats just a tall tale you tell in order to hide your desire to glorify yourself by (ridiculously) contributing content of no value that nobody will ever read to any replication crisis journal that will accept you

>>15824569
Dude, I'm a computer science PhD student. Except for shitty subfields like machine learning there is no replication crisis in CS.

>>15824608
CS isn't a science.

>>15824625
>he said while posting on an online forum with his computer

>>15771903
>than me, a mere SWE.
Im so sorry to hear that youre swedish.

>>15824643
Civil Engineering doesn't magically become a science when you drive on a bridge.

>>15824668
If a civil engineer invents a novel and innovative type of bridge, of course it is science.

>>15824682
Your inability to understand what science is isn't giving me any additional confidence that computer programming is a science.

>>15824372
If you actually knew some people in academia you would know that they fall mostly in these camps:
- overachieving midwits
- useless autistics
- "politicians"
None of these contribute much to their respective fields.
The 4th camp would be "the innovator" but those who fit the description are becoming exceedingly rare.

>>15824690
>The 4th camp would be "the innovator" but those who fit the description are becoming exceedingly rare.
In the sciences anyway. Most of them become engineers now because they're smart enough to know that engineers already made everything worthwhile.

>>15824684
I'm not a programmer, I'm researching novel algorithms.

>>15824714
So you're a mathematician.

>>15824716
No, I have a master's degree in CS.

>>15824721
And yet you just told me that you do mathematics research.

>>15824723
That's just hairsplitting.

>>15824729
The entire argument is about you falsely calling yourself a scientist when you don't use the scientific method.

>>15775210
taking 250 studies and 30 years to workout that a preposterous money priming psychological effect is nonsense is not a good advert for the scientific process

>>15824731
>hurr durr I know that you work as a researcher at a university
>hurr durr no, no, you are not a real scientist

>>15824741
Research != Science

>>15824744
This thread is a bottomless pit wrt. the average IQ of its posters.

>>15774749
In your dreams. It applies everywhere, even in motherfucking math.

>>15775210
The opposite. The point of science is reproducibility. Can't disprove something that can't be reproduced in the first place, it's easy enough to claim you dind't try to reproduce in good faith or that you don't know what you're doing and that's why you aren't getting the same results.

>>15787189
Same in biology for the same reasons by the way.

>>15798715
I have a PhD and there are no postdocs currently avaiable due to the state of the job market. I'll get my postdoc "soon enough". But holy hell is academia fucked up. It's cute how you have such high hopes to join in and all but believe me, this will permanently ruin your life. 90% of PhDs think doing a PhD was the worst mistake in their life for a reason. The rest are overwhelmingly silver spoon kids.

>>15813337
Why would they be paid per review instead of per hour or lump sum?

>>15824608
CS is very unique in this. Not only does it nearly not have a replication crisis (there are many cases of replication issues, and ML used to be just as good as anything CS until it became popular, in any case CS replication issues are genuinely rare compared to anything else), it also has next to no politics, which is unique across academic fields.
I wish I never left CS.

>>15824692
There hasn't been innovation in industry in the past 25 years.

>>15825721
I did a lot of machine learning stuff for my master's thesis. That is why my professor usually assigns ML papers he is requested to review to me. So far I have reviewed roughly a dozen ML papers. Mostly about deep learning. The quality is so low that it is laughable.
Only two of those papers contained all the information that is necessary to replicate the results, i.e., to train the model and check if the mentioned values of the metrics are correct. One paper had a table containing the types of hidden layers and the used hyper parameters. The other paper just contained a link to a github repository with the complete tensor flow source code used for training. Those two papers received an accept score from me. All the other received a hard reject. There is no way to check if the results of the rejected papers are accurate. The authors could have just forged them.
Unfortunately, you can find a lot of such low quality papers published in peer reviewed journals. Especially in open access journals that require the authors to pay an article processing charge.

>>15825723
>what are smartphones
>what are main stream electric cars
>what is actual usable satellite Internet

>>15826121
>Mostly about deep learning. The quality is so low that it is laughable.
Yup.
Since around 2016 it's been like this for everything even remotely related to Dl.
The problem is that the quality is no lower than in non-CS fields.
The worst is that, llike you say, this crap passes peer review now because all the conferences got subverted by big corps and act as recruiting events now, not scientific outlets. But there's nothing else.
The papers you passed almost certainly got rejected in the end.

>>15826144
Not only ahistorical but also clinically retarded.
Amazing.

>>15826514
>of course we live exactly like 25 years ago

>>15774780
it also affects the hard sciences - physics, chemistry, and biology; basically anything that involves long-ass and expensive experiments.
i don't know what it is like for the math fields presumably because they just use powerful computers to do their math experiments. is it like that for maths?

>>15826607
No, we live infinitely worse.
Go be underaged somewhere else, child.

>>15826641
We were not talking about the general quality of life, but innovations. There is certainly no lack of the latter.

>>15771903
Doubting your beliefs is a good thing. It's the starting point to having a better understanding of the world you live in and is central to an actually scientific understanding of the world you live in.

>>15826735
Not sure if retarded or merely pretending.jpg

>>15798715
A lot certainly are, it became especially apparent during the whole LK-99 hype

>>15771903
>SWE
>free economy
Lmao the software boom of the past 15 years has been propped up by ZIRP, its why the left coast is dying now.

>>15826848
Why not both?
Now enlighten me!

>>15826862
Smartphones were invented in the early 90's. The blackberry (late 90's) had more features than your iFail could ever dream to have.
You can't even tell the difference between repackaged slop from the 70's and something new, what "enlightenment" could you possibly accept or receive? You think something suddenly becoming popular means innovation. Maybe marketing innovation lmao. Electric cars are 3 years younger than the first ICE.

>>15826891
>The blackberry
A PDA is not a smartphone

>Electric cars are 3 years younger than the first ICE.
That is why I wrote *main stream* electric cars.

>>15826907
>blackberry
>a pda
>not a smartphone
You must be 18 or older to browse this site

>>15826861
>its why the left coast is dying now.
immigration is why California died

>>15827547
And the the end of low interest rates is going to rape the corpse.

>>15827547
California is the lowest IQ state, all because of immigration

>>15826144
>electric cars
they're worse than real cars, they're not an innovation

>>15829725
In fact electric cars were created before real cars, and real cars are the progressed form.

>>15811860
It’s a dumb meme making fun of dark matter and dark energy. It applies nowhere else, calm down, there are no people making up numbers wildly different to fit conclusions, at least unless they are paid to do that. And with dark matter and energy it was just some mitwit highschooler making this meme when they watched some video on YouTube about the topic.

>>15775210
>Hot take: a reproducibility crisis should expected; it's proof the scientific method is working
Congratulations for winning the dumbest fucking /sci/ take of 2023

>>15829738
dark matter is clearly fake, you have to be low iq to be incapable of recognizing that.

>>15830518
Smartest poster on /sci/!

>>15829728
Karl Benz invented the automobile in 1886, it was gas powered

>>15816008
beyond over

>>15832051
Nearly 100% over

>>15829728
>>15832007
global warming shills will lie about anything and everything
>why do we have a fraud crisis in science

>>15824682
thats engineering.

>>15831250
thanks

Academia is a meme my guy. And the more you approach theoretical sciences, the more this is apparent.
>be neurobiologist
>1st year in the lab, doing basic research on epilepsy
>get approached by a professor that asks me if he can use a part of the brains from my lab animals that I won't be using for his schizophrenia study
>confused, thought these were epilepsy models
>turns out they're also schizophrenia models. and alzheimer models. and adhd models, and depression models. the same fucking animals
I also know people who did adhd research on fucking fruit flies, i shit you not. complete circus

>>15835334
sounds like your whole life is just a big waste of time, how come you don't kys? what keeps you going through the motions?

>>15832990
>atheists will lie about anything and everything
Christians are the only group which believes that dishonesty is a sin. Thats why science worked when it was a strictly Christian affair and thats why science becomes toxic when Christians are excluded and atheists are allowed to take over

>>15771903
The good thing about knowing that all scientific publications are all fake is that you can now disregard them entirely and ignore all the idiot academics who publish all the lies

>>15828949
The bussy will inherit the earth. All flows within the currents of causality.

>>15834082
scientists are always feel compelled to take credit for others' achievements as a means of justifying their existence because they have none of their own to boast about

>>15790677

>>15795609
you cant be serious about this anon
>t.compbio working in a physiology campus
nature itself is split into their own sets of bullshit journals, not even getting into how methods being shipped in their own fucking journal now.
Our CS dept doesn't even use the same library proxy so ACM is isolated to one campus. The number of times I've ever gotten shit back from an ILLiad loan request for any topic was ONCE, and I'm hired as an RA, not a student in this dept anymore.
I'd have to leave my lab and physically drive to my undergrad campus to access one of many journals I can get for free on scihub. This uni probably spends a good quarter mil on journals (considering how much they pay my boss yearly), and I can't even access basic shit that's free elsewhere.

>>15789943
>le heckin' CHRISTIANS are all paragons of TRUTH and HONESTY, just like the chief hecker himself, Jeebus X!

>>15839703
based

>>15839703
>Our CS dept doesn't even use the same library proxy so ACM is isolated to one campus.
If the IT administrators of your university are too retarded to set up a proper campus network with VPN access, that is not the fault of publishers.

>>15840622
this, vpn access to campus networks is important because I like it

>>15812058
>No one is getting paid to do reviews
Of course not, its the vanity publishing industry, its stuff nobody wants to read. Doing reviews for garbage like that it a retarded pastime, don't you have anything more productive to do with your life?

Since there is no penalty for publishing lies in the scientific press, why should anyone trust anything published in the scientific press?

>>15771903
Ornately-crafted bait, but it's an interesting topic so whatever.

>Maybe you academics aren't actually changing the world, at least not the majority of you.
You'd be right; each one puts one brick on. Not a single one actually changed much at all. And yet at the end there's a building standing.

>>15784719
>computer science
>The replication crisis doesn't concern these as much
A surprising proportion of CS researchers won't post their code even though it can all be uploaded to github with 4 commands and takes like 5 seconds. They're afraid someone will dig into it and find a bug that forces them to retract.

I've seen a storage research team rush to publish a paper on a new technology, then turn around and publish another paper talking about the errors that made the first paper invalid. A self-licking ice cream cone of citations.

I've seen a paper about LLM where the researchers used another LLM to grade the results. So basically they made the score of LLM #2 their benchmark and then published when they got some improvement. Total garbage.

Papers are nothing but a stupid game. I made a mistake starting a PhD

>>15843469
>A surprising proportion of CS researchers won't post their code
If you plan to write a follow-up paper (or at least consider to do it), it might be a good idea to not publish your source code to lower the chance that someone else steals your idea.

>>15782971
The amount of people who seem to think Academia isn't corrupt is absurd. They need funding and to get funding, you definitely have to be testing something that won't piss the people giving you money off.

>>15784659
That's part or the problem. Anyone who actually cares either leaves or is in the process of fighting an uphill battle. The Science is never settled and stuff changes all the time. I think it's funny how people write off stuff all the time without even considering that it's a possibility. That doesn't mean assume it's true and act based off of that information, but it could be true. Most of the "the science is settled" people would have criticized popular scientists in the past because everyone thought they were wrong until they weren't. In 100 years we might learn that what we think is true right now, isn't,

>>15788857
It becomes a matter of engineer this study such that we get x results rather then let's test y to see if x happens

>>15794389
That's ironic because liberals aren't having children so it's null and void. Assuming they are having sex, they aren't having children, so what difference does it make? Neither is reproducing. One is just jacking off with extra steps.

>>15823508
A lot is environmental. Early childhood. You're literally a new human learning about the world and how to survive in it. You adapt to your environment and then you're no longer in the environment you grew up in. Now you have mental problems.

>>15835334
I get the idea
You breed a mouse to show excessive signs of some genetic or environmental disease. Mouse speedruns itself to the debilitation.
You can then test vs generations of the mice, allowing a lot of rapid tasting.

But like
What if you don't fundamentally understand the debilitation?
Doesn't that mean you can't fundamentally breed or infect mice for the debilitation and study?

>>15782971
>The problem might not be peer review... It's corruption.
peer review is corruption
"you have to believe my soiyence lies because these other soiyence liars also say the lies are true"

>>15845454
This. Without peer review it would be harder to enforce a corrupt environment because of open publication.

>>15845458
You can always self-publish your research. You don't have to use a renown journal.

>>15845904
What's the point of shouting in the void when you used to be able to present research in person before your nation's foremost scientific minds?

>>15845919
>present research in person before your nation's foremost scientific minds
The research which was presented at such occasions was checked by other scientists before, too. The process wasn't called "peer review" but even back then not any retard could just present his obviously incorrect bullshit in front of an expert audience.

>>15845454
>>15845458
>>15845919
People here feel suppressed by the scientific community because it won't allow them to publish a paper in which they call everyone racial slurs.

>>15845930
It was checked after, actually. It was called replication, and it was once a part of the scientific method.

>>15845942
You can still check a paper afterwards today. The peer review is just a pre-filter to get rid of the papers that are obviously bullshit and not worth the time and cost to publish and present them.

>>15845964
How can you believe in this idealistic view of "peer review" when the majority of published papers are provably misleading or fabricated? If you believe the system works this way, shouldn't it... Actually work?

>>15845966
There would be even more misleading or fabricated papers being published without the peer review system. Of course it is not perfect, but it is a step in the correct direction. It filters out the easy to spot bullshit papers.

>>15845969
Before peer review more disruptive innovations and accurate scientific advancements were made. Ever since it was implemented in the 1970s scientific advancement has slowed to a crawl and disruptive innovations don't occur except in industry.

>>15845971
Maybe all the easy discoveries have already been made?

>disruptive innovations don't occur except in industry
If you need ten of billions of USD to develop some new medication or CPU, of course only large companies are able to finance that. No single person or university is able to do that.

>>15845981
"Peer review" promotes mediocrity.

>>15845988
t. college dropout

>>15845991
Nope lol

>>15845988
Absolutely

>>15845981
The companies don't finance it either, their government money is just bigger.
But we should be focused on methodology if we can't afford the actual product, which they keep the how under wraps.

>>15845971
>disruptive innovations don't occur
because the midwit peer reviewers, who are fundamentally incapable of comprehending the thought processes of the tiny minority of highly intelligent scientists, prevent disruptive innovations
thats the nature of the peer review system, it relies on large numbers of reviewers and its statistically impossible to gather large numbers of high iq reviewers, the majority will always be midwits

>>15847865
>I'm so smart
>everyone else is dumb

>>15848415
Take time to go over uploads so they are perfect. Perfecting happens automatically in process. It gives you the choice to go on. Perfections supply the field.

>>15845988
Peer review is the monkeys on typewriters trying to produce Hamlet

>>15843469
https://neildegrassetyson.com/cv/#papers

>>15848772
At least they are trying instead of wasting their time shitposting on this board.

>>15849393
>t. wasting time shitposting on this board.
how come you have such a glaring lack of self awareness? do you have a severe personality disorder or something?

What exactly is it that was failing in being reproduced? Because if they included all the retarded dissertations in that figure, that's not a surprise.

>>15850645
>What exactly is it that was failing in being reproduced?
all of the false claims and forged data

>>15845991
>t. mediocrity

>>15772945
FPBP

>>15848415
Thanks for noticing how much smarter I am than you

>>15830449
Shut up please? He has a point, science is becoming a religion. They should all just kill themselves.

>>15856002
I’ve never thought of it that way. I guess you’re right.

>>15771903
One of the biggest bugs left unchecked in academia are the nazis that need the evidence to justify every retarded thing they can think of.

>>15771903
I discussed this in my web seminar (do you need evidence?). Science whentality. The performer of the experiement. Will use differenct callbacks when you take host of the body. No?

>>15773351
If youre dumbass cant tell if a picture is real blow your fucking brains out already

>>15771903
Science-1 should be focused into one subject matter at a time and I suggest Big Oil be the first topic they tackle

>>15856036
Ran the entire model and they still think its them

>>15771903
Dumb kids getting their ideas stolen. Poor kids

I broke my quantum computer can I get another one?

>>15826514
Are we going to ignore the fact that you need fresh eyes; perhaps dumb ones.

>>15771903
Last relevant paper I read was information being talked about as the fifth state of matter, I don’t care what an elite thinks they’re doing.

>>15845904
you can't, if its not in the peer reviewed journals then its not considered important and will be ignored

'Replication crisis' reveals modern research filled with false findings

https://azdailysun.com/news/nation-world/replication-crisis-in-psychology-research/article_43bf7e8a-6638-58f6-ac42-cbc736d17c10.html

>>15858133
>'Replication crisis' reveals modern research filled with false findings
> reveals
anyone who was even half paying attention already was well aware that all scientific publications are fake and gay

>>15825704
proof? How you fuck do you fail to replicate a proof?