5 ms·

Data never justifies anything by itself. The article is about what they did with the data to get the conclusions. Where do you think they went wrong?

pvaldes 5y ago
"Evidence of something" means "I can prove something with this data and I'm 100% sure about that this happened". The phrase "Evidence that Vikings lived in North America by at least the year 2021" is not justified by the data shown in the article. There are several assumptions scattered on the paper that are taken as evidence but are simply wishful thinking. For example: "All the three samples had the same age. Statistics say that this wouldn't be probable so is not salvaged wood and a man should have chopped it. Evidence" Using your math bodyguard to support something that can be disproved easily with logic or empirical observation is putting style over substance in science. It happens all that time... and is deeply wrong. Do you want to know what I would call go to a place and finding 100 trees that died in exactly the same day of the same year? I would call it a flood. A wildfire. A plague. A storm with strong winds. Botanical knowledge and observation of the real world teach us that mature trees die often in clusters by natural causes and that when a tree dies their neighbors get more exposed and increase its probability to fall in a domino effect also. So the authors assume (wrongly) that the samples are independent and end with a weak conclusion. Wishful thinking. This is just one of the problems that I spotted in the article and not even the worse offender I'm not saying that vikings weren't in America in 2021. I'm saying that when you scratch a little the article does not prove that. I'm sure that this year will be token for granted as true without critical thinking from now on because, hum... fits in the narrative. The problem probably is that some external information is simply taken for granted by specialists on this field (and I'm not one of them), so we can't evaluate it. It happens often.
- perl4ever 5y ago
  >"Evidence of something" means "I can prove something with this data and I'm 100% sure about that this happened". I see. A lot of people are comfortable with the idea that knowledge is always incomplete, and satisfied when they can get reasonably narrow error bars. I am not criticizing your way of looking at things directly, but I think it is a fact that normal science is done under a paradigm of less than absolute certainty. When there was an uncertainty of hundreds of years, and someone has an analysis that produces a very high probability of one specific year, then I think that is considered a meaningful accomplishment, and about as precise as can be expected. >I'm not saying that vikings weren't in America in 2021. I'm saying that when you scratch a little the article does not prove that. It sounds like you are saying that Vikings might not have been in America before Columbus. This does sound like you are going beyond rejecting this article, and rejects a major premise of the research in it. I don't believe this article was intended to establish that Europeans were there, only to narrow down the timeframe. If this article was completely wrong in their analysis, I don't think it would indicate to other scientists that there were no Europeans in the area before Columbus. Just that the uncertainty about when it was remains relatively large.
  - pvaldes 5y ago
    > It sounds like you are saying that Vikings might not have been in America before Columbus. Please don't use the straw man fallacy with me. Is very obvious that this is not what I'm saying.
    - perl4ever 5y ago
      Please don't tell me what is obvious to me. Instead, assume good faith and clarify. You said "I'm not saying the Vikings weren't..." which means to me that you are saying they may have and they may not have. Both are logically equivalent. In turn, that means that you seem (to me) to be asserting it is the burden of this article in Nature to demonstrate they were. My point was, that's a premise not something they are establishing. I believe they were just narrowing down the date. Therefore if they in fact totally failed at their job, it should not affect the question you bring up. It does not seem relevant that they didn't prove it, if they didn't set out to and it is a generally accepted fact. Does that clarify? I make plenty of mistakes, but I also read as carefully as I can.
  - pvaldes 5y ago
    Oh, I see where could be the problem. I just realized that I wrote 2021 when I wanted to write (obviously) 1021. My bad.
    - perl4ever 5y ago
      I interpreted it that way though, that was not the misunderstanding.
- torstenvl 5y ago
  > "Evidence of something" means "I can prove something with this data and I'm 100% sure about that this happened". No. Evidence is anything relevant and probative. Probative is not the same as dispositive. And only fools are "100% sure" of anything about the exterior world.
  - pvaldes 5y ago
    This is legal jargon. We are not in a legal context so is useless. Evidence in science has a different meaning, Is something that proves or refutes and hypothesis. You can't refute a scientific hypothesis but only a little, is a boolean term. An hypothesis in science is either true or false. Period. The equivalent terms for true or false in law (illegal or not) are much more elastic and adopt the shape of money.
    - torstenvl 5y ago
      It is not jargon. It is not domain-specific. Any fact that makes a conclusion more likely or less likely is evidence. It need not be conclusive to be evidence. For example, if we were to find a closed lipid membrane on Mars, or an increase in CO2 or methane or oxygen in its atmosphere, that would be evidence - but far from conclusive evidence - of life.
      - pvaldes 5y ago
        Nobody says "I have a proof of this, but is not really a proof-proof, just a feeling". That would be called "to make your results sexier for publishing purposes and deceiving your audience". An evidence validate an hypothesis that is just a question designed to be falsified, so is a type boolean. ¿Were vikings here in 1021? is boolean. The sum of answers add support to a theory, that isn't boolean stuff. We shouldn't confuse both terms.
        torstenvl 5y ago
        You appear to have a fundamental misunderstanding of the meaning of these words. Perhaps you are not a native speaker? In any case, I am afraid you are terribly mistaken, friend. Evidence remains evidence even when it is not enough to justify being "100% sure." Have a good evening.

pvaldes 5y ago

"Evidence of something" means "I can prove something with this data and I'm 100% sure about that this happened". The phrase "Evidence that Vikings lived in North America by at least the year 2021" is not justified by the data shown in the article. There are several assumptions scattered on the paper that are taken as evidence but are simply wishful thinking. For example: "All the three samples had the same age. Statistics say that this wouldn't be probable so is not salvaged wood and a man should have chopped it. Evidence" Using your math bodyguard to support something that can be disproved easily with logic or empirical observation is putting style over substance in science. It happens all that time... and is deeply wrong. Do you want to know what I would call go to a place and finding 100 trees that died in exactly the same day of the same year? I would call it a flood. A wildfire. A plague. A storm with strong winds. Botanical knowledge and observation of the real world teach us that mature trees die often in clusters by natural causes and that when a tree dies their neighbors get more exposed and increase its probability to fall in a domino effect also. So the authors assume (wrongly) that the samples are independent and end with a weak conclusion. Wishful thinking. This is just one of the problems that I spotted in the article and not even the worse offender I'm not saying that vikings weren't in America in 2021. I'm saying that when you scratch a little the article does not prove that. I'm sure that this year will be token for granted as true without critical thinking from now on because, hum... fits in the narrative. The problem probably is that some external information is simply taken for granted by specialists on this field (and I'm not one of them), so we can't evaluate it. It happens often.

perl4ever 5y ago
>"Evidence of something" means "I can prove something with this data and I'm 100% sure about that this happened". I see. A lot of people are comfortable with the idea that knowledge is always incomplete, and satisfied when they can get reasonably narrow error bars. I am not criticizing your way of looking at things directly, but I think it is a fact that normal science is done under a paradigm of less than absolute certainty. When there was an uncertainty of hundreds of years, and someone has an analysis that produces a very high probability of one specific year, then I think that is considered a meaningful accomplishment, and about as precise as can be expected. >I'm not saying that vikings weren't in America in 2021. I'm saying that when you scratch a little the article does not prove that. It sounds like you are saying that Vikings might not have been in America before Columbus. This does sound like you are going beyond rejecting this article, and rejects a major premise of the research in it. I don't believe this article was intended to establish that Europeans were there, only to narrow down the timeframe. If this article was completely wrong in their analysis, I don't think it would indicate to other scientists that there were no Europeans in the area before Columbus. Just that the uncertainty about when it was remains relatively large.
- pvaldes 5y ago
  > It sounds like you are saying that Vikings might not have been in America before Columbus. Please don't use the straw man fallacy with me. Is very obvious that this is not what I'm saying.
  - perl4ever 5y ago
    Please don't tell me what is obvious to me. Instead, assume good faith and clarify. You said "I'm not saying the Vikings weren't..." which means to me that you are saying they may have and they may not have. Both are logically equivalent. In turn, that means that you seem (to me) to be asserting it is the burden of this article in Nature to demonstrate they were. My point was, that's a premise not something they are establishing. I believe they were just narrowing down the date. Therefore if they in fact totally failed at their job, it should not affect the question you bring up. It does not seem relevant that they didn't prove it, if they didn't set out to and it is a generally accepted fact. Does that clarify? I make plenty of mistakes, but I also read as carefully as I can.
- pvaldes 5y ago
  Oh, I see where could be the problem. I just realized that I wrote 2021 when I wanted to write (obviously) 1021. My bad.
  - perl4ever 5y ago
    I interpreted it that way though, that was not the misunderstanding.
torstenvl 5y ago
> "Evidence of something" means "I can prove something with this data and I'm 100% sure about that this happened". No. Evidence is anything relevant and probative. Probative is not the same as dispositive. And only fools are "100% sure" of anything about the exterior world.
- pvaldes 5y ago
  This is legal jargon. We are not in a legal context so is useless. Evidence in science has a different meaning, Is something that proves or refutes and hypothesis. You can't refute a scientific hypothesis but only a little, is a boolean term. An hypothesis in science is either true or false. Period. The equivalent terms for true or false in law (illegal or not) are much more elastic and adopt the shape of money.
  - torstenvl 5y ago
    It is not jargon. It is not domain-specific. Any fact that makes a conclusion more likely or less likely is evidence. It need not be conclusive to be evidence. For example, if we were to find a closed lipid membrane on Mars, or an increase in CO2 or methane or oxygen in its atmosphere, that would be evidence - but far from conclusive evidence - of life.
    - pvaldes 5y ago
      Nobody says "I have a proof of this, but is not really a proof-proof, just a feeling". That would be called "to make your results sexier for publishing purposes and deceiving your audience". An evidence validate an hypothesis that is just a question designed to be falsified, so is a type boolean. ¿Were vikings here in 1021? is boolean. The sum of answers add support to a theory, that isn't boolean stuff. We shouldn't confuse both terms.
      - torstenvl 5y ago
        You appear to have a fundamental misunderstanding of the meaning of these words. Perhaps you are not a native speaker? In any case, I am afraid you are terribly mistaken, friend. Evidence remains evidence even when it is not enough to justify being "100% sure." Have a good evening.