The Unstoppable Momentum of Outdated Science • Hacker News Board

Re: The Unstoppable Momentum of Outdated Science

#21

Post by H_Pylori » Mon, Mar 08, 2021, 10:49 PM UTC

ravenstine wrote: post #10

This reminds me of a talk I watched about research into human pheromones. The speaker talked about how literature and studies on pheromones all lead back to 1 or 2 old studies that were basically considered bunk, but researchers kept citing it, and other researchers would also cite the other studies that linked back to the original low-quality studies. His point was that the whole field of research into human pheromo…

I call it "The founder effect" of bad studies. Essentially the first few studies set the tone for the rest of the field, and it's very hard to correct the course due to the momentum. I'm glad people are starting to notice.

Re: The Unstoppable Momentum of Outdated Science

#22

Post by quercusa » Mon, Mar 08, 2021, 10:56 PM UTC

zqfm wrote: post #2

Maybe off the wall, but my first thought is certificate revocation. Could there be (or is there already) a semi-centralized database of research that should no longer be cited? With maybe a dependency graph of research that maybe needs to be revisited?

There's something quite like this in US law. If you are citing a case in a legal brief, you need to make sure it hasn't been overturned along the way, so you 'Shepardize' it. From Wikipedia [0]:

Shepard's Citations is a citator used in United States legal research that provides a list of all the authorities citing a particular case, statute, or other legal authority. The verb Shepardizing refers to the process of consulting Shepard's to see if a case has been overturned, reaffirmed, questioned, or cited by later cases.

[0]: https://en.wikipedia.org/wiki/Shepard%27s_Citations

Re: The Unstoppable Momentum of Outdated Science

#23

Post by ncmncm » Mon, Mar 08, 2021, 10:59 PM UTC

carrolldunham wrote: post #12

before you spend 5 minutes searching for the real emissions line on the graph, it looks like a slight bolding of one of the labelled lines that ends in about 2018

The blue one with the dip around 2009?

Re: The Unstoppable Momentum of Outdated Science

#24

Post by marcus_holmes » Mon, Mar 08, 2021, 10:59 PM UTC

kjrose wrote: post #14

From my time working within an academic environment, I can attest this is absolutely true. Especially since the gatekeepers (older professors, etc) have a general interest to see these older articles cited more and referenced. It's not malicious, it's just a fact of how these things work. It's also why the general public's knowledge of how science works needs to be adjusted. Science is not a "100% correct" or "100% n…

I think we need to make the clear distinction between "Science" - the application of the Scientific Method, and "Academia" - the set of institutions, processes, and people who should be applying the scientific method, but are actually just trying to make a living in a messed-up system like the rest of us.

Science is always right in the long term.

Academia can be very wrong for a very long time.

Re: The Unstoppable Momentum of Outdated Science

#25

Post by SubiculumCode » Mon, Mar 08, 2021, 11:00 PM UTC

zqfm wrote: post #2

Maybe off the wall, but my first thought is certificate revocation. Could there be (or is there already) a semi-centralized database of research that should no longer be cited? With maybe a dependency graph of research that maybe needs to be revisited?

building that in Zotero would be great. I all ready like that they flag retracted papers.

Re: The Unstoppable Momentum of Outdated Science

#26

Post by pdonis » Mon, Mar 08, 2021, 11:00 PM UTC

incompatible wrote: post #11

philipkglass wrote: post #3

The graph shows emissions trajectories projected by the most commonly used climate scenarios (called SSP5-8.5 and RCP8.5, with labels on the right vertical axis), along with other scenario trajectories. Actual emissions to date (dark purple curve) and those of near-term energy outlooks (labeled as EIA, BP and ExxonMobil) all can be found at the very low end of the scenario range, and far below the most commonly used…

Looking a CO2 measurements, I don't see much sign that the rate of increase is slowing down: https://climate.nasa.gov/vital-signs/carbon-dioxide/ . Do the current models assume it will accelerate even faster?

The article is talking about annual CO2 emissions, not total CO2 levels. CO2 levels and their rate of increase are a lagging indicator; it takes many years for a reduction in emissions to show up in CO2 levels.

The models all assume that CO2 emissions will continue to increase; the main difference in the assumptions is how fast emissions are assumed to increase.

Re: The Unstoppable Momentum of Outdated Science

#27

Post by TaupeRanger » Mon, Mar 08, 2021, 11:01 PM UTC

How many times do we have to rediscover Kuhn's 'The Structure of Scientific Revolutions' and write articles about ideas that were well understood 60 years ago?

Re: The Unstoppable Momentum of Outdated Science

#28

Post by Shugarl » Mon, Mar 08, 2021, 11:08 PM UTC

TaupeRanger wrote: post #27

How many times do we have to rediscover Kuhn's 'The Structure of Scientific Revolutions' and write articles about ideas that were well understood 60 years ago?

Because there will always be people who just happened to never come across those ideas.

Re: The Unstoppable Momentum of Outdated Science

#29

Post by TaupeRanger » Mon, Mar 08, 2021, 11:09 PM UTC

Shugarl wrote: post #28

TaupeRanger wrote: post #27

How many times do we have to rediscover Kuhn's 'The Structure of Scientific Revolutions' and write articles about ideas that were well understood 60 years ago?

Because there will always be people who just happened to never come across those ideas.

I didn't ask "why", I asked how many times.

Re: The Unstoppable Momentum of Outdated Science

#30

Post by worldsayshi » Mon, Mar 08, 2021, 11:12 PM UTC

probably_wrong wrote: post #4

zqfm wrote: post #2

Maybe off the wall, but my first thought is certificate revocation. Could there be (or is there already) a semi-centralized database of research that should no longer be cited? With maybe a dependency graph of research that maybe needs to be revisited?

I've thinking how to make this idea work, but the more I think about it the less I think it's viable. First, you would be centralizing the ability to "cancel" papers. One bad member of the board could wreak havok. Simultaneously, people in charge of the database would be constantly bombarded by pleas (not all of them honest) to revoke this and that researcher's paper. So it would be bad for both sides. You could get…

You point out the flaws of one potential implementation of such system and reject the whole concept? I think it's conceivable that there is an implementation of such concept that would work better than not having it.

Example off the top of my head: Don't have one board decide which papers are cancelled. Have multiple. Have git repos with lists of cancelled papers. Have a process for finding which papers are flawed due to citing cancelled papers from a particular paper. Have browser extensions that tells you if the paper you're opening is carrying such flaws. Allow user of such extension to pick which board they trust. Have the extension link to a discussion about why the paper is cancelled. Give ability to challenge such claim in the open. Etc etc.

I'm sure many more such ideas can be found. The problem is probably rather funding of such system or if whoever funds it would have the right incentives to begin with.