Soft Science Gives Under Pressure
-
Hi, it's me, the guy who actually RTFAs. I think the core of the issue is summed up in this quote (from someone disappointed with the new result):
If the replication showed us anything, Baumeister says, it’s that the field has gotten hung up on computer-based investigations. “In the olden days there was a craft to running an experiment. You worked with people, and got them into the right psychological state and then measured the consequences. There’s a wish now to have everything be automated so it can be done quickly and easily online.” These days, he continues, there’s less and less actual behavior in the science of behavior. “It’s just sitting at a computer and doing readings.”
I mean, the original experiment was putting a bowl of freshly baked cookies next to a bowl of radishes, telling one group they could only eat cookies another they could only eat radishes, and then seeing how quick they would be to give up on a task, but that was rejected for this study as too hard to replicate widely, so instead of the cookies thing, they got themlooking at words on a screen and pressing a button if the letter e occurred in the right place. Which, to be fair, had been published as showing the same effect, but you'vegot to admit that disproving one of those doesn't necessarily disprove the other.
So there's a paradigm shift between smaller, more subjective studies, and larger more automated ones. I'm not in a position to take sides, which is good because I don'tknow which side I prefer. On the one hand, I get the importance of replicatability, on the other hand, removing the human element from psychology does seem kind of counterproductive. On the third hand, I've read enough pomo to know that sitting at a computer pressing a button while tryingto avoid entering any kind of emotional state represents like 99% of post-industrial life, so
<@xaade>Psychology isn'tgoing to advance until the world government makes their24/7 video surveillance footage publicly available./@xaade
Anyway, bottom line is that if you want to see someone who alreadyknows what conclusion they want to draw grabbing the first piece of evidence they see as final confirmation of that conclusion, look no further than the op.
-
@Buddy said in Soft Science Gives Under Pressure:
I mean, the original experiment
Yeah, I read that too, and someone needs to hand Baumeister his
, regardless of anything else.
-
@bb36e said in Soft Science Gives Under Pressure:
Related: here's a look into Amazon's nifty dropdown menu implementation
I like how he says it was discovered then forgotten. It was never forgotten. Some developers are just fucking incompetent and never bothered attempting to learn from the past.
-
@blakeyrat said in Soft Science Gives Under Pressure:
Some developers are just fucking incompetent and never bothered attempting to learn from the past.
Some? Most. It's one of the worst parts of this industry. :(
-
@blakeyrat said in Soft Science Gives Under Pressure:
Some developers are just fucking incompetent and never bothered attempting to learn from the past
Sounds like it being forgotten to me
-
@dkf said in Soft Science Gives Under Pressure:
Some? Most. It's one of the worst parts of this industry.
We like to forget about past inventions and then re-invent them badly and claim they're something new. HCI is not the only field where that happens. Just ask some random developer when virtual machines were invented. I guarantee that 90% of the answers will be at least 2 decades off.
-
@asdf I remember machine emulators were in use since early 70s. Maybe even later 60s. I cannot pinpoint the exact name of some dude or gal who came up with the idea, and I'm too lazy to google, but it seems like an obvious idea: you write an emulator, it tells you if your architecture and instruction set even makes sense in the long run, and development of OS, compilers and other software can proceed without having to wait for the actual hardware to appear.
Also, I don't think the LISP machine's processor actually had a LISP interpreter wired in the hardware... Correct me if I'm wrong.
-
@wft The oldest one I know of is the one that UCSD Pascal used, but I'm sure it wasn't the first. Anyway, two decades before Java.
-
@asdf said in Soft Science Gives Under Pressure:
I'm sure it wasn't the first
Wasn't the first even on the Apple II. Wozniak implemented a virtual 16-bit 16-register CPU called Sweet16 to make 16-bit ints less fiddly to work with on the 6502.
But all that is way too late for the earliest real-world VM implementations, which are at least as old as I am (early Sixties).
I think there's even a good argument to be made that the universal Turing machine - a Turing machine implemented in software on another Turing machine - properly counts as a VM, and Turing invented that in the late Thirties.
-
@flabdablet said in Soft Science Gives Under Pressure:
I think there's even a good argument to be made that the universal Turing machine - a Turing machine implemented in software on another Turing machine - properly counts as a VM, and Turing invented that in the late Thirties.
I wouldn't go as far as counting the mathematical construct as an actual implementation.
-
@asdf Hm. I'm sure it's not the oldest, but I know Windows 95 ran DOS and Windows 3.11 drivers (if needed) in a 16-bit VM.
Which is funny too, because virtually everybody everywhere used that VM. Microsoft just never bragged about having it, it ran as invisibly as possible.
-
@flabdablet said in Soft Science Gives Under Pressure:
But all that is way too late for the earliest real-world VM implementations, which are at least as old as I am (early Sixties).
The earliest practical ones were probably running on IBM mainframes. A lot of the most interesting stuff was invented back then by IBM, and it's unpatented now precisely because it was patented then and the patent has been expired for decades…
-
@AyGeePlus said in Soft Science Gives Under Pressure:
Science advances by moving from 'kinda right' to 'mostly right'.
More like "less wrong."
-
@blek said in Soft Science Gives Under Pressure:
Sure, but I don't see many people claiming that software development has anything to do with science.
Most software degrees are in Computer Science.
-
@Fox most actual software development has very little to do with computer science. Citation: TDWTF is still a thing after 12 years.
-
@Arantor said in Soft Science Gives Under Pressure:
@Fox most actual software development has very little to do with computer science. Citation: TDWTF is still a thing after 12 years.
True, but most degrees are still called Computer Science degrees.
-
@Fox yes, and?
Actual development is frequently carried out by people without degrees and it rarely seems to impact on development when it is done by someone with a degree.
Just because they have a brain, it does not mean they are using it.
Just because they have a degree, it does not mean they are using it.
-
@Arantor Plenty of fools with degrees, but real dangerous to assume that having a degree makes a person a fool. The way I look at it, for people who can code worth
and who are wanting to learn more about what they're doing, a degree is a reasonable way into the more complicated end of things. It's not impossible to learn everything without going to university, but the university forces students to learn the awkward bits that they usually ignore; there are things that are hard, but which let you do awesome things once you've mastered them.
-
@dkf said in Soft Science Gives Under Pressure:
but the university forces students to learn the awkward bits that they usually ignore;
Does it?
Back when I went, it didn't teach development methodologies, using source control, debugging techniques, barely touched on UI design, etc. I learned virtually nothing useful.
@dkf said in Soft Science Gives Under Pressure:
there are things that are hard, but which let you do awesome things once you've mastered them.
Oh good, then everybody else has to suffer with that horrible buggy "clever" code.
-
-
@dkf A needle pulling thread. Me, a name I call myself. Fa, a long long way to run!
-
@dkf said in Soft Science Gives Under Pressure:
It's not impossible to learn everything without going to university, but the university forces students to learn the awkward bits that they usually ignore
Not only it is possible, it is the only way at the end
Higher educations
- Should teach you the patience to learn (no StackOverflow copypasta instant gratification)
- Should teach you how to teach yourself, because no good professor chews up the subject for your consumption
- Should teach you some basic tools, like algebra, algorithms, probability and statistics
-
@dse Yes, but the basic tools might be different ones. If you've not touched algebra and statistics before going to university, you
have had your time wastedmust be following the Common Core…
-
@dkf Algebra that includes matrix decomposition and 3D geometric manifolds and PDE, probability that talks about random processes and bayesian inference. I am talking about higher education's take on what one should have learned its basics in the high-school.
-
@dse I think I did PDEs and random processes in (equivalent of) high school.
-
@dkf You must be Russian, or east European
-
@dse No, but I did take the Further Maths course that they were offering back then.
-
@Buddy said in Soft Science Gives Under Pressure:
Hi, it's me, the guy who actually RTFAs.
No, you're not and I'm not either, that's the thing.
I still don't have a clue about whether ego depletion exists or not. I tried to read the Slate article but fuck! -- it's a Slate article. It's awful. There's too much fluff and representing heroes as either the bad or good guys. I mean, why do I have to have a fable-like written story of one guy's life just to get to the meat of article? Because this guy was contracted for 2000-words article and he had to deliver? Fuck this shit, I give up.
But hey, I'm sure I saw @AyGeePlus linking to the article (paper) itself, this should be interesting. So I scroll up (and lose my fucking reading position, because NodeBB), I click and what do I see? The original ego depletion paper.
Wait, what?
Why?
Why did @AyGeePlus think it's important? Why did he think that linking to the OP is more important that linking to the actual recent study that is supposed to have debunked this original one? For lulz? Hey?
I mean, here is the guy who has made up his mind based on a fucking Slate cover story. A story written by a journalist. To be sold. Ergo, to catch attention, not to be correct. He read it and he believed it. And then he took the pain to find a link to a research paper, but not the one that's actually interesting.
So hey, yeah, it well might be that all of those ego depletion-related studies are bogus, it wouldn't be the first time that psychology was wrong. But it's not be judged based on what @AyGeePlus wants us to judge it and the conclusion should not be "psychology is not a science" or "psychology is all bogus" or even "psychology is mostly bogus". Psychology is an important science and we should fucking first to know that, as @Yamikuronue has said: UI/UX research is essentially psychology, usability as a whole is psychology and fuck, what do I know, it seems that architects and town planners actually find it helpful. Or this whole field of work environment-related studies that ensure that we all have at least almost-OK (right, @Weng?) working conditions. That's all bogus, too?
What I do hate is close-minded morons who jump to conclusions about whole fields/subjects based on a fucking small part of it being proven wrong, or as in this case: him having read an article that said they're wrong. (I mean, it cites interested parties, so it must be right, right?)
Hey, just realized I was originally replying to @Buddy, so I'll finish this way to: +1!
-
@boomzilla said in Soft Science Gives Under Pressure:
@AyGeePlus said in Soft Science Gives Under Pressure:
Science advances by moving from 'kinda right' to 'mostly right'.
More like "less wrong."
Funny you should say that:
-
@kt_ said in Soft Science Gives Under Pressure:
And then he took the pain to find a link to a research paper, but not the one that's actually interesting.
Maybe his ego got too depleted for him to pick the right link?
-
@kt_ I found a link to the pdf from the recent work page of the guy who blogged about being ‘in a very dark place’. I don't think its been published yet. https://static1.squarespace.com/static/550b09eae4b0147d03eda40d/t/56f6d84a0442629cf99e78a5/1459017803049/A-multi-lab-pre-registered-replication-of-the-ego-depletion-effect.pdf
Also, I found the page where the preliminary results were hosted pretty interesting. https://osf.io/92dhr/wiki/home/ Apparently this Registered Replication Report stuff is hipster github science
-
@AyGeePlus If you think the reproducibility problem is special or unique to psychology, you've got another think coming. The problem affects all of experimental academia, because careers a built on original research, not on "proving someone else's work for them" which is how a lot of academics view reproduction. After all, if someone wins a Nobel prize, the guys who reproduced it and helped prove it ain't gonna get shit.
This is a problem that impacts all of science, not just psychology and soft sciences. The fact that only psychology has suffered a crisis about it is what's absurd.
-
For people looking for a short summary, SciShow did a video on this issue:
-
Milk is good for you, milk causes cancer, milk prevents cancer...
Just eat the cows....
No, no, red meat is bad for you...
Red meat is good for you...
(some random guy): Your heart is red! Does it float.
That's right, red wine is good, so red meat must also be good.
(some other random guy): No.... my veganism! Must prove meat is bad
-
@antiquarian said in Soft Science Gives Under Pressure:
@dse said in Soft Science Gives Under Pressure:
And so do sociologists, political science and recent philosophers (not natural philosophy, or what later merged into real sciences and math). In fact most of humanities, and arts, all are not real science.
Who thinks philosophy is a science?
What exactly do you think "PhD" means? Remember, this is English, where lots of words gots lots of meanings.
-
White people might lose mental strength when they tried to talk about racial politics with a black scientist.
Fuck you article.... fuck you.
I just wanted to read an article about psychology, not get drawn into identity politics again.
It's like a fucking parasite on society.
@anyone who bitches about all threads ending up about politics in these forums.... this is why....
I can't fucking go get breakfast at McDonald's without the food menu insinuating that I'm a racist for not ordering the chicken items.
-
@kt_ said in Soft Science Gives Under Pressure:
conclusion should not be "psychology is not a science" or "psychology is all bogus" or even "psychology is mostly bogus". Psychology is an important science and we should fucking first to know that.
God no, we have to study the mind somehow. I agree completely.
My conclusion is that apparently a substantial subfield of psychology researchers are not doing science. And nobody noticed for years. On the other hand a third of papers in Nature are apparently irreplicable so wtf do I know.
-
@BaconBits said in Soft Science Gives Under Pressure:
the guys who reproduced it and helped prove it ain't gonna get shit
That's why it is traditionally a task given to the peons in grad school. It's a bit of a problem when it involves an instrument that costs more than the student's likely lifetime income, or when it involves an essentially impossible to reproduce event (good luck getting that exact hurricane to happen again!) but that just encourages some creativity.
-
@xaade One wonders how our great-grandparents survived long enough to have children.
-
@PleegWat by having lots of children, or by having running water and sewage available....
-
“I feel like the ground is moving from underneath me and I no longer know what is real and what is not.”
The things people put their faith in.... :(
-
Just leaving this here.
Even John Oliver chips in on academic research.
https://www.youtube.com/watch?v=0Rnq1NpHdmw
-
@Buddy said in Soft Science Gives Under Pressure:
I've read enough pomo
I read this as "porno".
@BaconBits said in Soft Science Gives Under Pressure:
The fact that only psychology has suffered a crisis about it is what's absurd.
It's not just psychology though. Here are a couple of posts from In The Pipeline on irreproducibility problems in the drug discovery industry (he has a bunch more, which you can look for if you're interested). I'd expect it's possible to find similar things across a range of disciplines, but I can't be bothered looking.
-
@Scarlet_Manuka said in Soft Science Gives Under Pressure:
irreproducibility problems in the drug discovery industry
It involves three things: bacteria, complicated purifications, and toxicity studies.
None of those usually include enough information to replicate them. It's the worst!
