5 ms·

While "I met a guy, didn't like him, I think this is why" is a bit too dismissive, your dismissal of a desire for factual data is disturbing. One person's expe

While "I met a guy, didn't like him, I think this is why" is a bit too dismissive, your dismissal of a desire for factual data is disturbing. One person's experience, no matter how broad for an individual, is still just one individual's anecdotal experience. This is true even if they have a sharp mind, decades of experience, and the best credentials. The principles of fact checking and bias analysis should not be relegated solely to academia.

coldtea 7y ago
>While "I met a guy, didn't like him, I think this is why" is a bit too dismissive, your dismissal of a desire for factual data is disturbing. I find the over-reliance on second hand ("factual") data, charts and figures, disturbing. People have to learn to observe, think, and understand themselves. Not just passively consume pre-made charts and statistics (which are the easiest thing to manipulate). And learning how to spot BS statistical claims wont help when the raw data can be themselves cherry-picked, manipulated, and diced in tons of ways. Not to mention that live experience is 360 (if one tries), where data will always paint less than the whole picture. One could arrange for great "official" charts and figures for every country -- and most countries do. Unless one gets on the field and talk to the people on the street and the workplace, and try to check the reality in various situations, they can have a totally BS picture painted for them by the statistics and "factual data". >One person's experience, no matter how broad for an individual, is still just one individual's anecdotal experience. Well, you're not 10000 people. You're just one, like everyone else is. In the end, whatever you're fed or read or watch, you have to make up your mind for yourself. >The principles of fact checking and bias analysis should not be relegated solely to academia. Nor should fact checking and bias analysis start and begin with data people are handed down from official or other sources. Those can range from perfectly accurate to badly compiled to totally and mischievously misleading (for saving face, incompetence, for profit, etc). If you lived in USSR, would you trust the official data, or you would try to balance things and do direct observation? You shouldn't blindly trust "facts" and "figures" anywhere else either...
- bovermyer 7y ago
  You keep saying charts and figures. Repeatedly, ad nauseum. As if that's what I was getting at. Quit talking about them, that's not relevant to this discussion. Comparing two data points will always yield better results than having just one. Data points don't have to be literal points on a graph. They can be discussions with people. Study historiography or journalism.
  - AnimalMuppet 7y ago
    You know, it's possible that Paul Graham might have had discussions with people, rather than just his own one-life anecdotal data...
    - bovermyer 7y ago
      No shit. Ever heard the phrase "trust, but verify?"
  - coldtea 7y ago
    >You keep saying charts and figures. Repeatedly, ad nauseum. Actually I've said the exact phrase twice in this subthread. And considering its the main subject of the subthread (whether those IMO are enough/mandatory or not) it's not really surprising. Perhaps you don't want like to hear it, so you're extra annoyed anytime I mention it, which makes you overcount? Maybe you need some charts and figures of how many times I said that expression. >Comparing two data points will always yield better results than having just one. Which is neither here nor there. A single person could collect 1000s of data points by observation, talking to different people, etc. You don't need a team to do that, or some official citation. There could be bias in said person? Sure. Same way the team/paper could have bias, be non-reproducible crap, be written to make some government body look good, be written to make someone money, etc. It's almost as is critical thinking by the person receiving the single person's anecdote or the, can I say it, charts and figures, is still required. >Study historiography or journalism. Oh, the irony.
TeMPOraL 7y ago
PG's essay isn't presenting itself as a scientific work presenting a theory. It's just one man's opinion. But 'arh68 above seems to be interpreting it as if it was such a scientific work.

coldtea 7y ago

>While "I met a guy, didn't like him, I think this is why" is a bit too dismissive, your dismissal of a desire for factual data is disturbing. I find the over-reliance on second hand ("factual") data, charts and figures, disturbing. People have to learn to observe, think, and understand themselves. Not just passively consume pre-made charts and statistics (which are the easiest thing to manipulate). And learning how to spot BS statistical claims wont help when the raw data can be themselves cherry-picked, manipulated, and diced in tons of ways. Not to mention that live experience is 360 (if one tries), where data will always paint less than the whole picture. One could arrange for great "official" charts and figures for every country -- and most countries do. Unless one gets on the field and talk to the people on the street and the workplace, and try to check the reality in various situations, they can have a totally BS picture painted for them by the statistics and "factual data". >One person's experience, no matter how broad for an individual, is still just one individual's anecdotal experience. Well, you're not 10000 people. You're just one, like everyone else is. In the end, whatever you're fed or read or watch, you have to make up your mind for yourself. >The principles of fact checking and bias analysis should not be relegated solely to academia. Nor should fact checking and bias analysis start and begin with data people are handed down from official or other sources. Those can range from perfectly accurate to badly compiled to totally and mischievously misleading (for saving face, incompetence, for profit, etc). If you lived in USSR, would you trust the official data, or you would try to balance things and do direct observation? You shouldn't blindly trust "facts" and "figures" anywhere else either...

bovermyer 7y ago
You keep saying charts and figures. Repeatedly, ad nauseum. As if that's what I was getting at. Quit talking about them, that's not relevant to this discussion. Comparing two data points will always yield better results than having just one. Data points don't have to be literal points on a graph. They can be discussions with people. Study historiography or journalism.
- AnimalMuppet 7y ago
  You know, it's possible that Paul Graham might have had discussions with people, rather than just his own one-life anecdotal data...
  - bovermyer 7y ago
    No shit. Ever heard the phrase "trust, but verify?"
- coldtea 7y ago
  >You keep saying charts and figures. Repeatedly, ad nauseum. Actually I've said the exact phrase twice in this subthread. And considering its the main subject of the subthread (whether those IMO are enough/mandatory or not) it's not really surprising. Perhaps you don't want like to hear it, so you're extra annoyed anytime I mention it, which makes you overcount? Maybe you need some charts and figures of how many times I said that expression. >Comparing two data points will always yield better results than having just one. Which is neither here nor there. A single person could collect 1000s of data points by observation, talking to different people, etc. You don't need a team to do that, or some official citation. There could be bias in said person? Sure. Same way the team/paper could have bias, be non-reproducible crap, be written to make some government body look good, be written to make someone money, etc. It's almost as is critical thinking by the person receiving the single person's anecdote or the, can I say it, charts and figures, is still required. >Study historiography or journalism. Oh, the irony.

TeMPOraL 7y ago

PG's essay isn't presenting itself as a scientific work presenting a theory. It's just one man's opinion. But 'arh68 above seems to be interpreting it as if it was such a scientific work.