[question] PF audience
14 views
Skip to first unread message
Elliot Temple
Apr 2, 2018, 4:10:17 PM4/2/18
to FI, FIGG
i started writing a new PF essay for the Critical Fallibilism website. the start is ok, then some of it is pretty rough and kinda written as notes. it’s also incomplete. i have a bunch of notes of good ideas and haven’t yet gone thru and decided which to include or not.
i think the problem is lack of clear audience. i had an idea how to start and that went ok. then later i kinda wrote to myself, said things that i would understand. that didn’t create text suitable for the website audience.
i don’t know what audience to target. it needs to be ppl who are good in some way. i don’t want to write targeting idiots. i don’t know how to successfully explain this to idiots. if i could explain it to the top 5% of ppl (even 1%), that’d be great. i’m not gonna worry about the bottom half.
i don’t want to target only Popperians or only Objectivists or only ppl who have read the whole FI website or anything that limiting.
but i don’t have a clear picture of what kind of merits, ability to understand things, goodness, etc, that i can expect the good audience to have. what resources like background knowledge, logical thinking, reading skill, etc. can i rely on?
the somewhat good ppl in the world vary a lot. idk what common themes i can work with. i don’t have that clearly defined enuf.
i need a clearer idea of what i think the audience already knows and needs to be told, what their problem situation is.
this relates to problems i commonly have with writing. i used to write lots of stuff for an audience of DD with little regard for whether other ppl understood. i also used to kinda imagine the world had lots of good, smart, reasonable, knowledgeable ppl i hadn’t found yet. but now i think at best i can expect to get some readers with maybe a few good traits, and a decent number of ok traits, rather than someone awesome. i dramatically lowered my opinion of what to expect from audiences and that’s problematic when i wanna write interested, somewhat advanced stuff that interests me (instead of super basic tutorials – which are what ppl need but they refuse to read those anyway cuz they are arrogant...)
one of my ideas i’ve used sometimes is to write for an imaginary smart young person (like aged 10, 15 or 20). another is to take a friend of mine and pretend they don’t know this specific topic at all, and maybe take some other pieces of knowledge away (like don’t assume they know CR or Oism), and then explain it in a way they’d be able to understand given their skill at reading, logical thinking, etc.
anyway, thoughts? help? i think this audience issue is very important.
also you can share thoughts on the specific text below if you want. plz don’t reply with like 2 typos and that’s it tho, i’d rather u just make it 100% clear you haven’t replied than do that since this isn’t even close to a finished product, i’ll be rewriting and editing everything myself anyway. but if u have a substantive comment about PF stuff, that’d be great.
# Paths Forward
*If I’m mistaken about this, how will I find out?*
A **path forward** is an answer to this question. It’s a way to make *progress* – a way to find out about and correct a *mistake*.
Paths forward are a major part of being rational because human beings are fallible. We can’t avoid making mistakes, we can’t guarantee ideas aren’t mistaken, and we commonly make mistakes.
Paths forward should appeal to anyone who understands fallibilism, and understands that knowledge creation works by an ::evolutionary:: process of generating ideas and correcting errors. It’s built on those ideas.
*If I’m mistaken about this, **and someone else understands the mistake and is willing to share a better idea**, how will I find out?*
Trying to find all your mistakes alone is a bad approach. You can correct lots of mistakes that way, but you can correct more mistakes if you get help from other people. Other people will think of ideas you don’t, and have different strengths and weaknesses than you. Even a person who is worse than you is probably better than you at something, and also probably lacks a weakness that you have. And even if you’re better than them at everything, you have limited time and energy, so you can benefit from ::comparative advantage:: (like in economics). A billion heads are better than one if you can find an effective way to collaborate to get useful information from them.
The internet allows us to receive criticism and suggestions from huge numbers of people. We should use it so that if someone wants to help us, they can.
If you try to figure out your mistakes and you miss some, that’s understandable. And if no one else knows about those mistakes, there’s no quick fix – just keep doing your best to use the methods of reason, and figure out approaches to life which are robust and resilient to error. But what if someone does know about a specific mistake you’re making, and they’d like to tell you, but they are powerless to help you because you block all their attempts to help? That’s inexcusable – and currently extremely common.
## Blocking Help Correcting Mistakes
How do people prevent themselves from receiving help? They don’t write down their ideas online, so potential helpers can’t read and comment. Or if they do write it online, they don’t engage with comments, or have comments disabled. If you don’t have online comments enabled on your writing, you need to be available for discussion in some other way – e.g. email – or people can’t help you. But people generally aren’t available for serious discussion by email or in any other medium. You can comment or email to tell people about a mistake in something they wrote and you don’t get any reply, or you maybe get a short, unproductive reply.
Helping people normally requires several back-and-forth communications. You need to ask some clarifying questions, get answers, and then you can tell them the mistake. You can’t just tell them the important mistake immediately because they wrote some ambiguous things (so you can initially just point out the ambiguity mistake, then you can point out the topical mistake once they clarify.) And you need to know what background knowledge someone has – e.g. are they familiar with Karl Popper? – before you know how to explain their mistake to them (what existing knowledge can you refer to in your explanation?). People prevent themselves from being helped by not making these back-and-forth communication steps available.
Another way people prevent being helped is they think a correction they get is wrong, so they ignore it. They don’t share the reason they think the correction is wrong, so that reason can’t be corrected. That gets things stuck. You need to publicly share your reason for rejecting each criticism of your position so that errors in that thinking can be corrected too, not just errors in the original.
When smart, knowledgeable people discuss, they normally have several layers of argument ready in advance. They have an initial argument, then depending on what you say they have their followup already prepared in advance. They also know their second and third followups for most common scenarios before the discussion even starts. But they can’t skip steps because which followup they use depends on what you say. At each step, they know several replies you might make and then what they’ll say next. To skip ahead, they’d have to list every reply you might say and all of their answers – it’d be really long and lots of work, and 90% of what they wrote wouldn’t be relevant to you.
What should happen when two smart, knowledgeable people debate is they quickly go through the initial steps. You say your standard opener, I say my standard reply, you say your standard followup, I say my standard followup, you say your standard followup, I say my standard followup, you say your standard followup, I say my standard followup, and things slow down and get interesting when someone says something the other guy isn’t already familiar with. But in practice this rarely happens. People usually get stuck before this. Why? Because they think the other guy is wrong and stop discussing before the point where he could say something new to them. They are like “I think that’s wrong and I already know 3 layers of followup arguments covering tons of different elaborations he might say. So I’ll just judge he’s wrong now to save time, and I will refuse to discuss.”
## Why People Block Paths Forward
People’s main excuse for blocking paths forward is they have limited time, and they don’t think it’s worthwhile to engage with would-be helpers *compared to* doing something else like reading a book where they might learn something. Many people offering criticism and suggestions are mistaken. Many smart, knowledgeable people get fed up with talking to idiots, and get cynical and pessimistic. So they mostly talk to people they know, people they are introduced to by people they know, people who got to the same in-person events as them, and people with public reputations. And they read writing from famous people, people who are approved of by gatekeepers (like journal editors who decide which articles to include), writing which receives positive reviews, etc. This approach blocks off tons of potential error correction. Reviewers, gatekeepers, reputation and social networking are all really unreliable ways to get the best info. They tend to block outliers, recent innovation (great ideas before they’ve become popular), unconventionalness/major-originality, and also just the vast majority of people (some small fraction of whom have something good to contribute right now).
There is a way to structure your thinking, learning, writing, etc., in order to solve this problem. If you develop your knowledge in the right way you can do much better. If you didn’t, you can transition.
## Conversations
A rational approach, compatible with paths forward, can address questions like these in conversations:
- This book/article/blog-post explains why you’re mistaken about that. Have you read it and written an answer to it it?
- If you haven’t looked at it, how do you know it’s mistaken and should be ignored? If you don’t know that, why are you disregarding it?
- Did someone else look at it, determine it’s mistaken, and write down an answer that you will endorse and take personal responsibility for?
- If your side of the debate has no public answers, aren’t you simply refusing to engage with contrary ideas?
- Are you refusing to consider my article because I don’t have some social status (fame, an intellectual reputation, the approval of an authority like a publisher or editor, etc.)? If it’s not that, what is it? If it is that, what do you expect me to do in the hypothetical scenario where I’m correct? Do you think I should get a PhD, network, suck up to people, etc, etc, as the required steps in order to be permitted to help you? Isn’t years of irrational torture too much to ask me to endure to participate in rational discussion? And wouldn’t you still ignore me even if i had a PhD, a published book, or whatever other credentials? I happen to know people who agree with me and have credentials, prestige, etc., but that isn’t enough for you. You selectively, biasedly accept credentials based on the people saying stuff you either agree with or find reasonable (doesn’t agree with you too much or in ways you don’t want to think about). But if someone has plenty of status/credentials/etc and you don’t want to listen, you just dismiss them as a crank or say that there are a lot of people with credentials and you’re busy. So really this is all just an excuse, and the broader fact is you don’t want to be corrected on some points, yeah? You make quick guesses about which ideas are any good and ignore the ones your initial judgement says is bad, and you block being corrected about that!
- If your answer is you’re too busy to consider everything and have to focus, will you be neutral on the issues you haven’t found time for? Accept your ignorance, stop making claims about them, stop taking sides – including in your own mind, not just publicly?
- Whatever your answer is for how helpers/critics can correct you, did you write it down? Did you tell them that? Did you expose it to criticism? Or is it all unwritten rules that you can change mid-criticism to suit your biases? You need stated policies for how you think, learn, address criticism, etc, to help prevent bias. (And the stated policies need to be real, clear, and followed. Often people’s stated policies are blatant lies containing false bragging, so readers don’t expect them to be true. If you’re serious about truth-seeking, you must have written policies that differentiate you in some way from a fraud. For example, if you say stuff about paths forward that would stand out as something different than the standard lip service to open-mindedness, reason, objectivity, etc. Another way to stand out would be to send money to everyone who has corrected you and to keep a list of corrections, with dates, and then people can see you often do listen, including recently, and genuinely value the help. And even if you rarely get successful corrections, you could still blog each unsuccessful correction you listened to and why you rejected it. People could then see how you handle input, criticize your handling of input, see that you do consider input in a reasonable way, learn from it, avoid repeating input you’ve heard before, etc.)
## How To Approach Knowledge
As you learn you should write down what you think are the good ideas that aren’t already written down by anyone, anywhere – including refutations of rival ideas. When you have new knowledge that isn’t yet part of the literature, that can’t yet be referenced, then it should be written down.
This simple rule means that whenever you’re questioned on your ideas, you can refer people to already written explanations. This makes it quick and easy to address tons and tons of inquiries from the public. You don’t have to write new answers when critics (would be helpers) try to correct you. You can just link them to existing material. You’ll only have to say something new if they say something new – at which point the issue is worth more attention (as long as its within your interests – if it’s not relevant to you so you’d rather work on something else, that’s fine, you can say you don’t know and you’re busy with other matters).
People often do the first step of this a decent amount. They write down their claims. But they don’t write down their answers to well known criticisms and counter-arguments in a decisive, canonical way. So they can’t say to critics, “That’s already answered, see here.” They have writing to reference to explain their positive claims, but not to address potential errors. This is because they have no clue about fallibilism and the importance of error correction, and it totally ruins their ability to use reason to learn.
## Libraries of Criticism
As you learn, you should learn more and more criticisms that can be re-used in many cases – the more general purpose the better. Then whenever someone says something, you will usually already have written down a criticism of it (or made note of a webpage or book where someone else did it) that you can refer to the person to. In the exceptional cases where something is new to you – not even in any of the many whole categories of ideas you already know criticisms of – then it’s worth attention.
The more you build up your library of pre-existing criticisms, the harder it is to create new ideas that aren’t immediately refuted by already-known criticisms. If you haven’t done this and find the public’s comments overwhelming, that’s on you – you don’t know much, so start learning! If you do this, then as you get more knowledgeable and popular, and get more inquiries and comments, you’ll also have a bigger library of that can address a much larger proportion of all communications to you so it still won’t take too long.
If you get really mega popular and busy cuz ur so great, you should by this time have some fans who will volunteer time or some money to hire help. What do you have them do? Simple. They can address inquiries by providing references to the already-written answers you would have used. They don’t need to be a genius or super great person or whatever to do that. So they can handle all the routine inquiries and pass on to you the small minority of inquiries with some new idea for you to consider. The best way to handle this is with a public forum, not privately, so that the public can see inquiries that were already answered, and can see that you actively answer inquiries (either personally or by proxy).
If you don’t set this kinda stuff up then you’re blocking paths forward. You will make mistakes, people in the public will know corrections, and you will stay wrong.
Currently only a handful of people do paths forward – all of them part of the FI discussion community. It’s why I have such a huge lead on other philosophers outside our discussion community.
Elliot Temple
www.elliottemple.com
i think the problem is lack of clear audience. i had an idea how to start and that went ok. then later i kinda wrote to myself, said things that i would understand. that didn’t create text suitable for the website audience.
i don’t know what audience to target. it needs to be ppl who are good in some way. i don’t want to write targeting idiots. i don’t know how to successfully explain this to idiots. if i could explain it to the top 5% of ppl (even 1%), that’d be great. i’m not gonna worry about the bottom half.
i don’t want to target only Popperians or only Objectivists or only ppl who have read the whole FI website or anything that limiting.
but i don’t have a clear picture of what kind of merits, ability to understand things, goodness, etc, that i can expect the good audience to have. what resources like background knowledge, logical thinking, reading skill, etc. can i rely on?
the somewhat good ppl in the world vary a lot. idk what common themes i can work with. i don’t have that clearly defined enuf.
i need a clearer idea of what i think the audience already knows and needs to be told, what their problem situation is.
this relates to problems i commonly have with writing. i used to write lots of stuff for an audience of DD with little regard for whether other ppl understood. i also used to kinda imagine the world had lots of good, smart, reasonable, knowledgeable ppl i hadn’t found yet. but now i think at best i can expect to get some readers with maybe a few good traits, and a decent number of ok traits, rather than someone awesome. i dramatically lowered my opinion of what to expect from audiences and that’s problematic when i wanna write interested, somewhat advanced stuff that interests me (instead of super basic tutorials – which are what ppl need but they refuse to read those anyway cuz they are arrogant...)
one of my ideas i’ve used sometimes is to write for an imaginary smart young person (like aged 10, 15 or 20). another is to take a friend of mine and pretend they don’t know this specific topic at all, and maybe take some other pieces of knowledge away (like don’t assume they know CR or Oism), and then explain it in a way they’d be able to understand given their skill at reading, logical thinking, etc.
anyway, thoughts? help? i think this audience issue is very important.
also you can share thoughts on the specific text below if you want. plz don’t reply with like 2 typos and that’s it tho, i’d rather u just make it 100% clear you haven’t replied than do that since this isn’t even close to a finished product, i’ll be rewriting and editing everything myself anyway. but if u have a substantive comment about PF stuff, that’d be great.
# Paths Forward
*If I’m mistaken about this, how will I find out?*
A **path forward** is an answer to this question. It’s a way to make *progress* – a way to find out about and correct a *mistake*.
Paths forward are a major part of being rational because human beings are fallible. We can’t avoid making mistakes, we can’t guarantee ideas aren’t mistaken, and we commonly make mistakes.
Paths forward should appeal to anyone who understands fallibilism, and understands that knowledge creation works by an ::evolutionary:: process of generating ideas and correcting errors. It’s built on those ideas.
*If I’m mistaken about this, **and someone else understands the mistake and is willing to share a better idea**, how will I find out?*
Trying to find all your mistakes alone is a bad approach. You can correct lots of mistakes that way, but you can correct more mistakes if you get help from other people. Other people will think of ideas you don’t, and have different strengths and weaknesses than you. Even a person who is worse than you is probably better than you at something, and also probably lacks a weakness that you have. And even if you’re better than them at everything, you have limited time and energy, so you can benefit from ::comparative advantage:: (like in economics). A billion heads are better than one if you can find an effective way to collaborate to get useful information from them.
The internet allows us to receive criticism and suggestions from huge numbers of people. We should use it so that if someone wants to help us, they can.
If you try to figure out your mistakes and you miss some, that’s understandable. And if no one else knows about those mistakes, there’s no quick fix – just keep doing your best to use the methods of reason, and figure out approaches to life which are robust and resilient to error. But what if someone does know about a specific mistake you’re making, and they’d like to tell you, but they are powerless to help you because you block all their attempts to help? That’s inexcusable – and currently extremely common.
## Blocking Help Correcting Mistakes
How do people prevent themselves from receiving help? They don’t write down their ideas online, so potential helpers can’t read and comment. Or if they do write it online, they don’t engage with comments, or have comments disabled. If you don’t have online comments enabled on your writing, you need to be available for discussion in some other way – e.g. email – or people can’t help you. But people generally aren’t available for serious discussion by email or in any other medium. You can comment or email to tell people about a mistake in something they wrote and you don’t get any reply, or you maybe get a short, unproductive reply.
Helping people normally requires several back-and-forth communications. You need to ask some clarifying questions, get answers, and then you can tell them the mistake. You can’t just tell them the important mistake immediately because they wrote some ambiguous things (so you can initially just point out the ambiguity mistake, then you can point out the topical mistake once they clarify.) And you need to know what background knowledge someone has – e.g. are they familiar with Karl Popper? – before you know how to explain their mistake to them (what existing knowledge can you refer to in your explanation?). People prevent themselves from being helped by not making these back-and-forth communication steps available.
Another way people prevent being helped is they think a correction they get is wrong, so they ignore it. They don’t share the reason they think the correction is wrong, so that reason can’t be corrected. That gets things stuck. You need to publicly share your reason for rejecting each criticism of your position so that errors in that thinking can be corrected too, not just errors in the original.
When smart, knowledgeable people discuss, they normally have several layers of argument ready in advance. They have an initial argument, then depending on what you say they have their followup already prepared in advance. They also know their second and third followups for most common scenarios before the discussion even starts. But they can’t skip steps because which followup they use depends on what you say. At each step, they know several replies you might make and then what they’ll say next. To skip ahead, they’d have to list every reply you might say and all of their answers – it’d be really long and lots of work, and 90% of what they wrote wouldn’t be relevant to you.
What should happen when two smart, knowledgeable people debate is they quickly go through the initial steps. You say your standard opener, I say my standard reply, you say your standard followup, I say my standard followup, you say your standard followup, I say my standard followup, you say your standard followup, I say my standard followup, and things slow down and get interesting when someone says something the other guy isn’t already familiar with. But in practice this rarely happens. People usually get stuck before this. Why? Because they think the other guy is wrong and stop discussing before the point where he could say something new to them. They are like “I think that’s wrong and I already know 3 layers of followup arguments covering tons of different elaborations he might say. So I’ll just judge he’s wrong now to save time, and I will refuse to discuss.”
## Why People Block Paths Forward
People’s main excuse for blocking paths forward is they have limited time, and they don’t think it’s worthwhile to engage with would-be helpers *compared to* doing something else like reading a book where they might learn something. Many people offering criticism and suggestions are mistaken. Many smart, knowledgeable people get fed up with talking to idiots, and get cynical and pessimistic. So they mostly talk to people they know, people they are introduced to by people they know, people who got to the same in-person events as them, and people with public reputations. And they read writing from famous people, people who are approved of by gatekeepers (like journal editors who decide which articles to include), writing which receives positive reviews, etc. This approach blocks off tons of potential error correction. Reviewers, gatekeepers, reputation and social networking are all really unreliable ways to get the best info. They tend to block outliers, recent innovation (great ideas before they’ve become popular), unconventionalness/major-originality, and also just the vast majority of people (some small fraction of whom have something good to contribute right now).
There is a way to structure your thinking, learning, writing, etc., in order to solve this problem. If you develop your knowledge in the right way you can do much better. If you didn’t, you can transition.
## Conversations
A rational approach, compatible with paths forward, can address questions like these in conversations:
- This book/article/blog-post explains why you’re mistaken about that. Have you read it and written an answer to it it?
- If you haven’t looked at it, how do you know it’s mistaken and should be ignored? If you don’t know that, why are you disregarding it?
- Did someone else look at it, determine it’s mistaken, and write down an answer that you will endorse and take personal responsibility for?
- If your side of the debate has no public answers, aren’t you simply refusing to engage with contrary ideas?
- Are you refusing to consider my article because I don’t have some social status (fame, an intellectual reputation, the approval of an authority like a publisher or editor, etc.)? If it’s not that, what is it? If it is that, what do you expect me to do in the hypothetical scenario where I’m correct? Do you think I should get a PhD, network, suck up to people, etc, etc, as the required steps in order to be permitted to help you? Isn’t years of irrational torture too much to ask me to endure to participate in rational discussion? And wouldn’t you still ignore me even if i had a PhD, a published book, or whatever other credentials? I happen to know people who agree with me and have credentials, prestige, etc., but that isn’t enough for you. You selectively, biasedly accept credentials based on the people saying stuff you either agree with or find reasonable (doesn’t agree with you too much or in ways you don’t want to think about). But if someone has plenty of status/credentials/etc and you don’t want to listen, you just dismiss them as a crank or say that there are a lot of people with credentials and you’re busy. So really this is all just an excuse, and the broader fact is you don’t want to be corrected on some points, yeah? You make quick guesses about which ideas are any good and ignore the ones your initial judgement says is bad, and you block being corrected about that!
- If your answer is you’re too busy to consider everything and have to focus, will you be neutral on the issues you haven’t found time for? Accept your ignorance, stop making claims about them, stop taking sides – including in your own mind, not just publicly?
- Whatever your answer is for how helpers/critics can correct you, did you write it down? Did you tell them that? Did you expose it to criticism? Or is it all unwritten rules that you can change mid-criticism to suit your biases? You need stated policies for how you think, learn, address criticism, etc, to help prevent bias. (And the stated policies need to be real, clear, and followed. Often people’s stated policies are blatant lies containing false bragging, so readers don’t expect them to be true. If you’re serious about truth-seeking, you must have written policies that differentiate you in some way from a fraud. For example, if you say stuff about paths forward that would stand out as something different than the standard lip service to open-mindedness, reason, objectivity, etc. Another way to stand out would be to send money to everyone who has corrected you and to keep a list of corrections, with dates, and then people can see you often do listen, including recently, and genuinely value the help. And even if you rarely get successful corrections, you could still blog each unsuccessful correction you listened to and why you rejected it. People could then see how you handle input, criticize your handling of input, see that you do consider input in a reasonable way, learn from it, avoid repeating input you’ve heard before, etc.)
## How To Approach Knowledge
As you learn you should write down what you think are the good ideas that aren’t already written down by anyone, anywhere – including refutations of rival ideas. When you have new knowledge that isn’t yet part of the literature, that can’t yet be referenced, then it should be written down.
This simple rule means that whenever you’re questioned on your ideas, you can refer people to already written explanations. This makes it quick and easy to address tons and tons of inquiries from the public. You don’t have to write new answers when critics (would be helpers) try to correct you. You can just link them to existing material. You’ll only have to say something new if they say something new – at which point the issue is worth more attention (as long as its within your interests – if it’s not relevant to you so you’d rather work on something else, that’s fine, you can say you don’t know and you’re busy with other matters).
People often do the first step of this a decent amount. They write down their claims. But they don’t write down their answers to well known criticisms and counter-arguments in a decisive, canonical way. So they can’t say to critics, “That’s already answered, see here.” They have writing to reference to explain their positive claims, but not to address potential errors. This is because they have no clue about fallibilism and the importance of error correction, and it totally ruins their ability to use reason to learn.
## Libraries of Criticism
As you learn, you should learn more and more criticisms that can be re-used in many cases – the more general purpose the better. Then whenever someone says something, you will usually already have written down a criticism of it (or made note of a webpage or book where someone else did it) that you can refer to the person to. In the exceptional cases where something is new to you – not even in any of the many whole categories of ideas you already know criticisms of – then it’s worth attention.
The more you build up your library of pre-existing criticisms, the harder it is to create new ideas that aren’t immediately refuted by already-known criticisms. If you haven’t done this and find the public’s comments overwhelming, that’s on you – you don’t know much, so start learning! If you do this, then as you get more knowledgeable and popular, and get more inquiries and comments, you’ll also have a bigger library of that can address a much larger proportion of all communications to you so it still won’t take too long.
If you get really mega popular and busy cuz ur so great, you should by this time have some fans who will volunteer time or some money to hire help. What do you have them do? Simple. They can address inquiries by providing references to the already-written answers you would have used. They don’t need to be a genius or super great person or whatever to do that. So they can handle all the routine inquiries and pass on to you the small minority of inquiries with some new idea for you to consider. The best way to handle this is with a public forum, not privately, so that the public can see inquiries that were already answered, and can see that you actively answer inquiries (either personally or by proxy).
If you don’t set this kinda stuff up then you’re blocking paths forward. You will make mistakes, people in the public will know corrections, and you will stay wrong.
Currently only a handful of people do paths forward – all of them part of the FI discussion community. It’s why I have such a huge lead on other philosophers outside our discussion community.
Elliot Temple
www.elliottemple.com
PAS
Apr 2, 2018, 8:08:53 PM4/2/18
to FIGG, FI
On Apr 2, 2018, at 1:10 PM, Elliot Temple <cu...@curi.us> wrote:
> i started writing a new PF essay for the Critical Fallibilism website. the start is ok, then some of it is pretty rough and kinda written as notes. it’s also incomplete. i have a bunch of notes of good ideas and haven’t yet gone thru and decided which to include or not.
>
> i think the problem is lack of clear audience. i had an idea how to start and that went ok. then later i kinda wrote to myself, said things that i would understand. that didn’t create text suitable for the website audience.
>
> i don’t know what audience to target. it needs to be ppl who are good in some way. i don’t want to write targeting idiots. i don’t know how to successfully explain this to idiots. if i could explain it to the top 5% of ppl (even 1%), that’d be great. i’m not gonna worry about the bottom half.
>
> i don’t want to target only Popperians or only Objectivists or only ppl who have read the whole FI website or anything that limiting.
>
> but i don’t have a clear picture of what kind of merits, ability to understand things, goodness, etc, that i can expect the good audience to have. what resources like background knowledge, logical thinking, reading skill, etc. can i rely on?
>
> the somewhat good ppl in the world vary a lot. idk what common themes i can work with. i don’t have that clearly defined enuf.
>
> i need a clearer idea of what i think the audience already knows and needs to be told, what their problem situation is.
I think you already know most or all of what I’m intending to write.
> i started writing a new PF essay for the Critical Fallibilism website. the start is ok, then some of it is pretty rough and kinda written as notes. it’s also incomplete. i have a bunch of notes of good ideas and haven’t yet gone thru and decided which to include or not.
>
> i think the problem is lack of clear audience. i had an idea how to start and that went ok. then later i kinda wrote to myself, said things that i would understand. that didn’t create text suitable for the website audience.
>
> i don’t know what audience to target. it needs to be ppl who are good in some way. i don’t want to write targeting idiots. i don’t know how to successfully explain this to idiots. if i could explain it to the top 5% of ppl (even 1%), that’d be great. i’m not gonna worry about the bottom half.
>
> i don’t want to target only Popperians or only Objectivists or only ppl who have read the whole FI website or anything that limiting.
>
> but i don’t have a clear picture of what kind of merits, ability to understand things, goodness, etc, that i can expect the good audience to have. what resources like background knowledge, logical thinking, reading skill, etc. can i rely on?
>
> the somewhat good ppl in the world vary a lot. idk what common themes i can work with. i don’t have that clearly defined enuf.
>
> i need a clearer idea of what i think the audience already knows and needs to be told, what their problem situation is.
I’m writing it anyway cuz:
- It is important and has affected lots of what I have done in my life. It probably will affect choices I make in the future.
- I might be wrong about some or all of it and learn something.
- You or someone else might not already know something I incorrectly think of as background.
I don’t know any way to reliably identify, let alone target top 1% or 5% people other than serious long-time FI regulars, which is more like .01%.
But I think getting directly to people in the top half is pretty easy. For that, target people who are voluntarily pursuing something requiring significant personal commitment and an intellectual component. Some examples ** not a complete list **:
* Active investors and entrepreneurs
* Homeschoolers
* Book authors (and authors of equivalent quality on-line material)
* People who hold formal or informal leadership roles in Conservative and Libertarian movements
* Programmers
* Ham radio operators
* Leaders of Christian or Jewish religious groups as well as Freethought / Atheist groups
* Self-defense oriented firearm and martial arts enthusiasts. These pursuits require thinking about and taking responsibility for high-impact safety and moral questions, and the culture around them is more serious than other “sports" as a result.
* Serious / competitive video gamers
I think the vast majority of people meeting at least one of the above (or other, similar activities) are top half.
Further, I think people who do 2 or more things like that at once are more likely than not (>50% chance) to be in the top 20%.
One way to target those people is at the intersections. For example, homeschoolers who write quality online blogs; self-defense enthusiasts who are also church leaders.
I don’t have a guess about how to target any better than top 20%. If you want to reach ~5 people in the top 1% the only way I know to do it is to reach ~100 people in the top 20%, which would mean reaching ~200 people with 2 or more intersecting top-half activities.
I think some differences for the top 20% vs. the rest of the population are:
- More forward thinking / future oriented
- Greater ability to understand and generalize from abstractions
- Greater capacity for persistence toward some chosen goal
- Better at reading, logic, and math
- Less fucked up personal lives (family, friends, finances)
I’d expect top 5% and top 1% to be even better than top 20% on these, but I don’t know how proportional the gains are.
Even at the top 1% I don’t think people have these traits at the level actually required to be successful at FI. They’ll need to make further progress on all of them. But they have a lot less distance to cover to reach competence than the rest.
What do they need to be told? If I had to pick one thing it’d be:
- You’re quite smart compared to most people. But you’re nowhere near as smart as you could be with already-known tools and methods that you’re not using.
I don’t know of a reliable way to communicate that message. If stated directly my guess is most of the top 1% will believe the first sentence and dismiss the second.
PAS
Elliot Temple
Apr 2, 2018, 9:57:36 PM4/2/18
to fallibl...@googlegroups.com, FI
You're talking about how to select and reach a particular audience. I'd call it marketing and outreach stuff. I'm talking about what audience to use mentally for making writing decisions. This is different. You say little that would be of use for making writing decisions. You do have some broad guidance about writing but not much more useful than "write for somewhat smart people, not dumb people".
maybe you mix these issues in your mind? i see them as separate. you can write for an audience without any particular way to reach them. you just write what they would like and then your readers can self-select by whether they like it (some ppl outside the target audience will like it, btw.). if ppl who like it share it with ppl similar to themselves, then your audience can find itself. the point here is marketing or having any way to reach the target audience (as against just putting it online for anyone) is not a necessary component of writing to target some problem situations and not others.
strategies for outreach and marketing have value but i wasn't thinking about that or planning to do much there. having a problem situation to address with writing is more important. to write publicly, i need some writing goals that work reasonably well for a decent amount of other ppl (as i think e.g. the FI website accomplishes). but i don't need readers in order to write.
i think there's some good things about writing for audiences other than yourself, like figuring out how to explain something to someone who doesn't already know. but there's difficulties, e.g. if you think some audiences suck too much to write for then you need some other idea about to aim for. (most ppl don't think much about audience when writing, but they are just going on unstated ideas about it. those ideas can be flawed and you can decide they aren't a good approach if you find out what you're doing.)
anyway, writing needs a purpose. your email is more about how to accomplish the purpose, but to write it the more important and prior step is to have some purpose in mind at all. you can plan to fill in other details, like demographics to advertise to, at a later date. what i want is some decision making strategies for reasonably canonical, future-proof writing, and i want some idea of who it might help and how which i can try to optimize.
Elliot Temple
www.elliottemple.com
PAS
Apr 2, 2018, 11:12:02 PM4/2/18
to fallibl...@googlegroups.com, FI
> You're talking about how to select and reach a particular audience. I'd call it marketing and outreach stuff. I'm talking about what audience to use mentally for making writing decisions. This is different. You say little that would be of use for making writing decisions. You do have some broad guidance about writing but not much more useful than "write for somewhat smart people, not dumb people”.
For example, if you want write about paths forward I intended to suggest that you consider how you would explain PF in terms that someone who’s a christian leader and self defense advocate would care about & understand.
Like suppose your audience is these guys:
http://www.warriortalk.com/forumdisplay.php?263-Warrior-Culture-Living-Well-and-Fighting-Well
You might not ever specifically try to bring what you write to their attention. But my guess is that having them in mind (and/or people at other intersections) would give you a specific enough idea of an audience to improve your writing.
> maybe you mix these issues in your mind? i see them as separate. you can write for an audience without any particular way to reach them. you just write what they would like and then your readers can self-select by whether they like it (some ppl outside the target audience will like it, btw.). if ppl who like it share it with ppl similar to themselves, then your audience can find itself. the point here is marketing or having any way to reach the target audience (as against just putting it online for anyone) is not a necessary component of writing to target some problem situations and not others.
Generally I think it’s good if those two are aligned. It’s good if you can write for an audience you have some ability to reach. But it's not a requirement.
> strategies for outreach and marketing have value but i wasn't thinking about that or planning to do much there. having a problem situation to address with writing is more important. to write publicly, i need some writing goals that work reasonably well for a decent amount of other ppl (as i think e.g. the FI website accomplishes). but i don't need readers in order to write.
>
> i think there's some good things about writing for audiences other than yourself, like figuring out how to explain something to someone who doesn't already know. but there's difficulties, e.g. if you think some audiences suck too much to write for then you need some other idea about to aim for. (most ppl don't think much about audience when writing, but they are just going on unstated ideas about it. those ideas can be flawed and you can decide they aren't a good approach if you find out what you're doing.)
>
> anyway, writing needs a purpose. your email is more about how to accomplish the purpose, but to write it the more important and prior step is to have some purpose in mind at all. you can plan to fill in other details, like demographics to advertise to, at a later date. what i want is some decision making strategies for reasonably canonical, future-proof writing, and i want some idea of who it might help and how which i can try to optimize.
PAS
Elliot Temple
Apr 3, 2018, 2:00:10 AM4/3/18
to FIGG, FI
> What do they need to be told? If I had to pick one thing it’d be:
> - You’re quite smart compared to most people. But you’re nowhere near as smart as you could be with already-known tools and methods that you’re not using.
>
> I don’t know of a reliable way to communicate that message. If stated directly my guess is most of the top 1% will believe the first sentence and dismiss the second.
I like this. I think it’s a good message.
> - You’re quite smart compared to most people. But you’re nowhere near as smart as you could be with already-known tools and methods that you’re not using.
>
> I don’t know of a reliable way to communicate that message. If stated directly my guess is most of the top 1% will believe the first sentence and dismiss the second.
I take it you now believe it. What convinced you? My first guess is the main thing was learning about the methods themselves (gradually over time) – guesses and criticism, fallibilism, evolution and error correction, critical discussion, rejecting intellectual authority, yesno, paths forward, powering up and doing easy/cheap things, not overreaching. That is, unfortunately, hard to condense.
What do others here think?
> You’re quite smart compared to most people. But you’re nowhere near as smart as you could be with already-known tools and methods that you’re not using.
Elliot Temple
www.fallibleideas.com
Elliot Temple
Apr 3, 2018, 10:47:04 PM4/3/18
to FIGG, FI
On Apr 2, 2018, at 11:00 PM, Elliot Temple <cu...@curi.us> wrote:
> On Apr 2, 2018, at 5:08 PM, PAS <p...@paipas.com> wrote:
>
>> What do they need to be told? If I had to pick one thing it’d be:
>> - You’re quite smart compared to most people. But you’re nowhere near as smart as you could be with already-known tools and methods that you’re not using.
>>
>> I don’t know of a reliable way to communicate that message. If stated directly my guess is most of the top 1% will believe the first sentence and dismiss the second.
>
> I like this. I think it’s a good message.
People often pretend the issue is they aren’t as smart (and knowledgeable) as me. That is an issue, but the even bigger issue is generally honesty.
> On Apr 2, 2018, at 5:08 PM, PAS <p...@paipas.com> wrote:
>
>> What do they need to be told? If I had to pick one thing it’d be:
>> - You’re quite smart compared to most people. But you’re nowhere near as smart as you could be with already-known tools and methods that you’re not using.
>>
>> I don’t know of a reliable way to communicate that message. If stated directly my guess is most of the top 1% will believe the first sentence and dismiss the second.
>
> I like this. I think it’s a good message.
PF helps keep you honest. How are you supposed to biasedly ignore the ideas you don’t like when you have to answer every idea? And when everything is publicly documented? PF gets intellectuals to follow a *structured approach* instead of winging it ... and that takes away tons of their opportunities to be biased.
Think about the people who refuse PF and ignore my criticisms. Is the issue they are too stupid and ignorant to understand my points? Not primarily. They don’t want to try that hard/seriously/honestly because of some kind of bias against me (e.g. cuz they guess I’m low prestige/status/popularity, or they know I’m associated with someone their peers dislike such as Ayn Rand or associated with some ideas their peers dislike such as rejecting induction).
Elliot Temple
www.curi.us
PAS
Apr 3, 2018, 10:47:24 PM4/3/18
to FIGG, FI
On Apr 2, 2018, at 11:00 PM, Elliot Temple <cu...@curi.us> wrote:
> On Apr 2, 2018, at 5:08 PM, PAS <p...@paipas.com> wrote:
>
>> What do they need to be told? If I had to pick one thing it’d be:
>> - You’re quite smart compared to most people. But you’re nowhere near as smart as you could be with already-known tools and methods that you’re not using.
>>
>> I don’t know of a reliable way to communicate that message. If stated directly my guess is most of the top 1% will believe the first sentence and dismiss the second.
>
> I like this. I think it’s a good message.
>
> I take it you now believe it.
I believe it at least enough to stay around rather than leave.
> On Apr 2, 2018, at 5:08 PM, PAS <p...@paipas.com> wrote:
>
>> What do they need to be told? If I had to pick one thing it’d be:
>> - You’re quite smart compared to most people. But you’re nowhere near as smart as you could be with already-known tools and methods that you’re not using.
>>
>> I don’t know of a reliable way to communicate that message. If stated directly my guess is most of the top 1% will believe the first sentence and dismiss the second.
>
> I like this. I think it’s a good message.
>
> I take it you now believe it.
> What convinced you? My first guess is the main thing was learning about the methods themselves (gradually over time) – guesses and criticism, fallibilism, evolution and error correction, critical discussion, rejecting intellectual authority, yesno, paths forward, powering up and doing easy/cheap things, not overreaching. That is, unfortunately, hard to condense.
But I think that’s OK. I don’t think I should expect to be thoroughly convinced given my current time+interest+skills in researching such topics.
What convinced me at least enough not to leave was approximately as you say (learning some of the methods). More specifically, learning a specific explanation for how to get knowledge was significant.
I had never heard of Popper but also was neither intellectually nor intuitively attached to the JTB model of knowledge. I picked up JTB culturally and it was probably part of the bad college philosophy class I deliberately forgot.
So something resembling JTB is probably what I would have regurgitated if pressed for how to get knowledge. But mostly I didn’t directly think about knowledge.
I incorrectly interpreted the suggestion that was made to read Popper’s Conjectures and Refutations. I thought it was suggested to read the online chapter here:
http://philosophyfaculty.ucsd.edu/faculty/rarneson/Courses/popperphil1.pdf
I later learned that what was actually suggested was to read the whole C&R book, here:
https://www.amazon.com/Conjectures-Refutations-Scientific-Knowledge-Routledge/dp/0415285941?tag=curi04-20
…which I wouldn’t have done (still haven’t) and so probably wouldn’t have worked on me.
The online chapter was short enough for me to successfully get through, was nothing like what I’d heard about knowledge before, and matched my inexplicit intuitions about what was going on. So I figured there might be more stuff like that worth sticking around to find out.
PAS
Elliot Temple
Apr 3, 2018, 10:58:57 PM4/3/18
to fallibl...@googlegroups.com, FI
On Apr 3, 2018, at 7:47 PM, PAS <p...@paipas.com> wrote:
> On Apr 2, 2018, at 11:00 PM, Elliot Temple <cu...@curi.us> wrote:
>
>> On Apr 2, 2018, at 5:08 PM, PAS <p...@paipas.com> wrote:
>>
>>> What do they need to be told? If I had to pick one thing it’d be:
>>> - You’re quite smart compared to most people. But you’re nowhere near as smart as you could be with already-known tools and methods that you’re not using.
>>>
>>> I don’t know of a reliable way to communicate that message. If stated directly my guess is most of the top 1% will believe the first sentence and dismiss the second.
>>
>> I like this. I think it’s a good message.
>>
>> I take it you now believe it.
>
> I believe it at least enough to stay around rather than leave.
>
>> What convinced you? My first guess is the main thing was learning about the methods themselves (gradually over time) – guesses and criticism, fallibilism, evolution and error correction, critical discussion, rejecting intellectual authority, yesno, paths forward, powering up and doing easy/cheap things, not overreaching. That is, unfortunately, hard to condense.
>
> I am not thoroughly convinced. For example, I’m not convinced that learnable ideas & methods account for approximately all differences in thinking ability between healthy humans, whereas genetics and diet and brain chemicals account for approximately none.
>
> But I think that’s OK. I don’t think I should expect to be thoroughly convinced given my current time+interest+skills in researching such topics.
This ability to step back to a different (meta) level and look at the situation with perspective *is itself a thinking skill which I’ve been teaching*. I think it’s really important and this application of it is reasonable in a way most people are not reasonable.
> On Apr 2, 2018, at 11:00 PM, Elliot Temple <cu...@curi.us> wrote:
>
>> On Apr 2, 2018, at 5:08 PM, PAS <p...@paipas.com> wrote:
>>
>>> What do they need to be told? If I had to pick one thing it’d be:
>>> - You’re quite smart compared to most people. But you’re nowhere near as smart as you could be with already-known tools and methods that you’re not using.
>>>
>>> I don’t know of a reliable way to communicate that message. If stated directly my guess is most of the top 1% will believe the first sentence and dismiss the second.
>>
>> I like this. I think it’s a good message.
>>
>> I take it you now believe it.
>
> I believe it at least enough to stay around rather than leave.
>
>> What convinced you? My first guess is the main thing was learning about the methods themselves (gradually over time) – guesses and criticism, fallibilism, evolution and error correction, critical discussion, rejecting intellectual authority, yesno, paths forward, powering up and doing easy/cheap things, not overreaching. That is, unfortunately, hard to condense.
>
> I am not thoroughly convinced. For example, I’m not convinced that learnable ideas & methods account for approximately all differences in thinking ability between healthy humans, whereas genetics and diet and brain chemicals account for approximately none.
>
> But I think that’s OK. I don’t think I should expect to be thoroughly convinced given my current time+interest+skills in researching such topics.
> What convinced me at least enough not to leave was approximately as you say (learning some of the methods). More specifically, learning a specific explanation for how to get knowledge was significant.
>
> I had never heard of Popper but also was neither intellectually nor intuitively attached to the JTB model of knowledge. I picked up JTB culturally and it was probably part of the bad college philosophy class I deliberately forgot.
>
> So something resembling JTB is probably what I would have regurgitated if pressed for how to get knowledge. But mostly I didn’t directly think about knowledge.
>
> I incorrectly interpreted the suggestion that was made to read Popper’s Conjectures and Refutations. I thought it was suggested to read the online chapter here:
> http://philosophyfaculty.ucsd.edu/faculty/rarneson/Courses/popperphil1.pdf
>
> I later learned that what was actually suggested was to read the whole C&R book, here:
> https://www.amazon.com/Conjectures-Refutations-Scientific-Knowledge-Routledge/dp/0415285941?tag=curi04-20
>
> …which I wouldn’t have done (still haven’t) and so probably wouldn’t have worked on me.
Popper chapters in most books are independent essays/lecture-transcripts so you can easily read them individually. and i list the best ones on the FI book list.
you don’t need to, and should not, wait until you feel like you have the time or interest to read a whole book. Popper presents his stuff in smaller chunks.
Elliot Temple
www.elliottemple.com
Alisa Zinov'yevna Rosenbaum
Apr 28, 2018, 5:47:49 AM4/28/18
to fallibl...@googlegroups.com, FI
On Mon, Apr 2, 2018 at 4:10 PM, Elliot Temple <cu...@curi.us> wrote:
> Another way to stand out would be to send money to everyone who has corrected you and to keep a list of corrections, with dates, and then people can see you often do listen, including recently, and genuinely value the help. And even if you rarely get successful corrections, you could still blog each unsuccessful correction you listened to and why you rejected it. People could then see how you handle input, criticize your handling of input, see that you do consider input in a reasonable way, learn from it, avoid repeating input you’ve heard before, etc.)
That's a smart idea.
> Another way to stand out would be to send money to everyone who has corrected you and to keep a list of corrections, with dates, and then people can see you often do listen, including recently, and genuinely value the help. And even if you rarely get successful corrections, you could still blog each unsuccessful correction you listened to and why you rejected it. People could then see how you handle input, criticize your handling of input, see that you do consider input in a reasonable way, learn from it, avoid repeating input you’ve heard before, etc.)
I could offer to send money, say, to the first finder of any mistake in anything one publishes. I wonder what a good payment amount policy would be? I wouldn't want it to be large enough discourage me from publishing at all. On the other hand, I would want it to be large enough to incentivize people. Currently I'm still making typos, so it's enough to incentivize typical readers; I don't have to incentivize only FI experts.
Other organizations, authors, and software developers offer bounties as well. For example:
Google offers cash payments of up to $31337 for reports of security holes in Google products [GOOG]:
> Rewards for qualifying bugs range from $100 to $31,337.
They publish a Hall of Fame of people who have "reported valid security bugs and helped us make Google products safer." [HOF]
Computer science author and software developer Donald Knuth has a pays for errors reported in his books or programs [KNU]:
>> In my books I offer rewards for the first person who finds any particular error... In software I similarly pay for errors in the TEX and METAFONT programs. The reward was doubling every year: It started out at $2.56, then it went to $5.12, and so on, until it reached $327.68, at which time I stopped doubling.
> There’s one man who lives near Frankfurt who would probably have more than $1,000 if he cashed all the checks I’ve sent him. There’s a man in Los Gatos, California, whom I’ve never met, who cashes a check for $2.56 about once a month, and that’s been going on for some years now. Altogether I’ve written more than 2,000 checks over the years, and the average amount exceeds $8.00 per check.
D. J. Bernstein has offered a bug bounty for his djbdns software since at least 2001. At that time the reward was $500 [DJB500]. It is currently $1000 [DJB1000].
> I offer $1000 to the first person to publicly report a verifiable security hole in the latest version of djbdns.
>
> My judgment is final as to what constitutes a security hole in djbdns. Any disputes will be reported here.
Saying "Any disputes will be reported here" is a good policy. It doesn't help D. J. Bernstein find out when he's wrong. But if he and a bug reporter ever disagree, readers of his page will at least be able to check out the dispute and decide for themselves.
[GOOG] Google. "Google Vulnerability Reward Program (VRP) Rules" https://www.google.com/about/appsecurity/reward-program/
[KNU] Donald Knuth (2002-03). "All Questions Answered" http://www.ams.org/notices/200203/fea-knuth.pdf
[DJB500] D. J. Bernstein (2001-02-03). "The djbdns security guarantee" https://web.archive.org/web/20010203230300/https://cr.yp.to/djbdns/guarantee.html
[DJB1000] D. J. Bernstein. "The djbdns security guarantee" https://cr.yp.to/djbdns/guarantee.html
[HOF] https://www.google.com/about/appsecurity/hall-of-fame/archive/
Alisa Zinov'yevna Rosenbaum
Apr 28, 2018, 5:53:46 AM4/28/18
to fallibl...@googlegroups.com, FI
On Mon, Apr 02, 2018 at 01:10:14PM -0700, Elliot Temple wrote:
> As you learn you should write down what you think are the good ideas that aren’t already written down by anyone, anywhere – including refutations of rival ideas. When you have new knowledge that isn’t yet part of the literature, that can’t yet be referenced, then it should be written down.
Yes, good idea. And if you learn something that IS part of the literature, post what you learned to FI along with a quote and a reference. People wouldn't run out of things to post to FI if they followed these tips.
> As you learn you should write down what you think are the good ideas that aren’t already written down by anyone, anywhere – including refutations of rival ideas. When you have new knowledge that isn’t yet part of the literature, that can’t yet be referenced, then it should be written down.
Elliot Temple
Apr 28, 2018, 1:36:37 PM4/28/18
to fallibl...@googlegroups.com, FI
> On Apr 28, 2018, at 2:47 AM, Alisa Zinov'yevna Rosenbaum <petrogradp...@gmail.com> wrote:
>
> On Mon, Apr 2, 2018 at 4:10 PM, Elliot Temple <cu...@curi.us> wrote:
>
>> Another way to stand out would be to send money to everyone who has corrected you and to keep a list of corrections, with dates, and then people can see you often do listen, including recently, and genuinely value the help. And even if you rarely get successful corrections, you could still blog each unsuccessful correction you listened to and why you rejected it. People could then see how you handle input, criticize your handling of input, see that you do consider input in a reasonable way, learn from it, avoid repeating input you’ve heard before, etc.)
>
> That's a smart idea.
>
> I could offer to send money, say, to the first finder of any mistake in anything one publishes. I wonder what a good payment amount policy would be? I wouldn't want it to be large enough discourage me from publishing at all. On the other hand, I would want it to be large enough to incentivize people. Currently I'm still making typos, so it's enough to incentivize typical readers; I don't have to incentivize only FI experts.
if you were counting FI posts as publishing then even $1/error would risk way more money, while simultaneously also possibly being too small for people to care much.
> Other organizations, authors, and software developers offer bounties as well. For example:
>
> Google offers cash payments of up to $31337 for reports of security holes in Google products [GOOG]:
>
>> Rewards for qualifying bugs range from $100 to $31,337.
>
> They publish a Hall of Fame of people who have "reported valid security bugs and helped us make Google products safer." [HOF]
Another example of payments for errors:
http://curi.us/2101-school-mistreated-edward-thorp
> The [chemistry] course was taught by a famous professor, and we were using his book. As he was then preparing a revision, he offered 10 cents per misprint to the first student to report it. I set to work and soon brought him a list of ten errors to see if he would pay. He gave me my dollar. Encouraged, I came back with a list of seventy-five more mistakes. That netted me $7.50 but he wasn’t happy. When I returned a few days later with several hundred he explained that they needed to be errors, not mere misprints. Despite my objections, he disqualified nearly all of them. This unilateral retroactive change in the deal, which I would later encounter often on Wall Street, done by someone for their benefit just because they could get away with it, violated my sense of fair play. I quit reporting additional corrections.
Elliot Temple
www.elliottemple.com
Alisa Zinov'yevna Rosenbaum
Apr 28, 2018, 2:25:32 PM4/28/18
to fallibl...@googlegroups.com, FI
On Sat, Apr 28, 2018 at 1:36 PM, Elliot Temple wrote:
>> On Apr 28, 2018, at 2:47 AM, Alisa Zinov'yevna Rosenbaum wrote:
>> On Mon, Apr 2, 2018 at 4:10 PM, Elliot Temple wrote:
>>> Another way to stand out would be to send money to everyone who has corrected you and to keep a list of corrections, with dates, and then people can see you often do listen, including recently, and genuinely value the help. And even if you rarely get successful corrections, you could still blog each unsuccessful correction you listened to and why you rejected it. People could then see how you handle input, criticize your handling of input, see that you do consider input in a reasonable way, learn from it, avoid repeating input you’ve heard before, etc.)
>>
>> That's a smart idea.
>>
>> I could offer to send money, say, to the first finder of any mistake in anything one publishes. I wonder what a good payment amount policy would be? I wouldn't want it to be large enough discourage me from publishing at all. On the other hand, I would want it to be large enough to incentivize people. Currently I'm still making typos, so it's enough to incentivize typical readers; I don't have to incentivize only FI experts.
>
> you could pay for first errors in blog posts only and then use a higher amount than you do with FI posts. if you write 5 posts per month on average, then $10/error (more than enough to get typos reported) would only risk $50/month.
There are three good ideas in what you just said that I hadn't considered:
- Paying only for the *first error* in a given piece of writing. I was thinking of paying for the *first report of any given error*. Suppose I publish a piece of writing with 5 different errors. With the way I was thinking about it, I would risk having to make 5 payments. With your idea, I would only risk having to make 1 payment.
- Calculating the rate at which I would be risking money by offering an error bounty.
- Having different error bounties for different categories of writing. I knew that Google pays different amounts for security vulnerabilities in different products, but I didn't apply that idea to my problem.
> if you were counting FI posts as publishing then even $1/error would risk way more money, while simultaneously also possibly being too small for people to care much.
Yeah.
>> Other organizations, authors, and software developers offer bounties as well. For example:
>>
>> Google offers cash payments of up to $31337 for reports of security holes in Google products [GOOG]:
>>
>>> Rewards for qualifying bugs range from $100 to $31,337.
>>
>> They publish a Hall of Fame of people who have "reported valid security bugs and helped us make Google products safer." [HOF]
>
> Hall of "Fame"? lol. Hall of super underpaid programmers. 31k is so low as a max reward when they'd have to pay a developer something like that to look for a security hole for a month, and he could easily find zero security holes that month, or only a minor one.
Maintaining a "Hall of Fame" lets Google pay less than they would otherwise pay for the same amount of security work. Google is a prestigious company, and some independent security researchers like being listed in Google's "Hall of Fame".
> why would you cap out around a month of pay for the biggest issues that could affect millions of your customers?
Yeah. I don't understand why Google pays so little for such important problem reports.
Wikipedia has an article on the "market rate for zero-day exploits": https://en.wikipedia.org/wiki/Market_for_zero-day_exploits
Zerodium is a company that buys security vulnerabilities and resells them to their corporate and government clients. Zerodium's motto is "We pay BIG bounties, not bug bounties". Zerodium claims to pay "up to $1.5 million" for an iPhone "Remote Jailbreak with Persistence" (http://zerodium.com/program.html). The fine print on that page says:
> All payouts are subject to change without notice, at the discretion of ZERODIUM.
That's not good. But Zerodium says that someone who finds a security vulnerability can get a written "pre-offer" from Zerodium without disclosing the details of what they found.
https://zerodium.com/faq.html :
> To receive a pre-offer for your research, you can submit minimal details about your exploit (without submitting the exploit). ZERODIUM will evaluate the minimal details, and will eventually confirm its interest and send you a pre-offer.
> Another example of payments for errors:
>
> http://curi.us/2101-school-mistreated-edward-thorp
>
>> The [chemistry] course was taught by a famous professor, and we were using his book. As he was then preparing a revision, he offered 10 cents per misprint to the first student to report it. I set to work and soon brought him a list of ten errors to see if he would pay. He gave me my dollar. Encouraged, I came back with a list of seventy-five more mistakes. That netted me $7.50 but he wasn’t happy. When I returned a few days later with several hundred he explained that they needed to be errors, not mere misprints. Despite my objections, he disqualified nearly all of them. This unilateral retroactive change in the deal, which I would later encounter often on Wall Street, done by someone for their benefit just because they could get away with it, violated my sense of fair play. I quit reporting additional corrections.
Good example of how *not* to run an error bounty program.
>> On Apr 28, 2018, at 2:47 AM, Alisa Zinov'yevna Rosenbaum wrote:
>> On Mon, Apr 2, 2018 at 4:10 PM, Elliot Temple wrote:
>>> Another way to stand out would be to send money to everyone who has corrected you and to keep a list of corrections, with dates, and then people can see you often do listen, including recently, and genuinely value the help. And even if you rarely get successful corrections, you could still blog each unsuccessful correction you listened to and why you rejected it. People could then see how you handle input, criticize your handling of input, see that you do consider input in a reasonable way, learn from it, avoid repeating input you’ve heard before, etc.)
>>
>> That's a smart idea.
>>
>> I could offer to send money, say, to the first finder of any mistake in anything one publishes. I wonder what a good payment amount policy would be? I wouldn't want it to be large enough discourage me from publishing at all. On the other hand, I would want it to be large enough to incentivize people. Currently I'm still making typos, so it's enough to incentivize typical readers; I don't have to incentivize only FI experts.
>
> you could pay for first errors in blog posts only and then use a higher amount than you do with FI posts. if you write 5 posts per month on average, then $10/error (more than enough to get typos reported) would only risk $50/month.
- Paying only for the *first error* in a given piece of writing. I was thinking of paying for the *first report of any given error*. Suppose I publish a piece of writing with 5 different errors. With the way I was thinking about it, I would risk having to make 5 payments. With your idea, I would only risk having to make 1 payment.
- Calculating the rate at which I would be risking money by offering an error bounty.
- Having different error bounties for different categories of writing. I knew that Google pays different amounts for security vulnerabilities in different products, but I didn't apply that idea to my problem.
> if you were counting FI posts as publishing then even $1/error would risk way more money, while simultaneously also possibly being too small for people to care much.
>> Other organizations, authors, and software developers offer bounties as well. For example:
>>
>> Google offers cash payments of up to $31337 for reports of security holes in Google products [GOOG]:
>>
>>> Rewards for qualifying bugs range from $100 to $31,337.
>>
>> They publish a Hall of Fame of people who have "reported valid security bugs and helped us make Google products safer." [HOF]
>
> Hall of "Fame"? lol. Hall of super underpaid programmers. 31k is so low as a max reward when they'd have to pay a developer something like that to look for a security hole for a month, and he could easily find zero security holes that month, or only a minor one.
> why would you cap out around a month of pay for the biggest issues that could affect millions of your customers?
Wikipedia has an article on the "market rate for zero-day exploits": https://en.wikipedia.org/wiki/Market_for_zero-day_exploits
Zerodium is a company that buys security vulnerabilities and resells them to their corporate and government clients. Zerodium's motto is "We pay BIG bounties, not bug bounties". Zerodium claims to pay "up to $1.5 million" for an iPhone "Remote Jailbreak with Persistence" (http://zerodium.com/program.html). The fine print on that page says:
> All payouts are subject to change without notice, at the discretion of ZERODIUM.
That's not good. But Zerodium says that someone who finds a security vulnerability can get a written "pre-offer" from Zerodium without disclosing the details of what they found.
https://zerodium.com/faq.html :
> To receive a pre-offer for your research, you can submit minimal details about your exploit (without submitting the exploit). ZERODIUM will evaluate the minimal details, and will eventually confirm its interest and send you a pre-offer.
> Another example of payments for errors:
>
> http://curi.us/2101-school-mistreated-edward-thorp
>
>> The [chemistry] course was taught by a famous professor, and we were using his book. As he was then preparing a revision, he offered 10 cents per misprint to the first student to report it. I set to work and soon brought him a list of ten errors to see if he would pay. He gave me my dollar. Encouraged, I came back with a list of seventy-five more mistakes. That netted me $7.50 but he wasn’t happy. When I returned a few days later with several hundred he explained that they needed to be errors, not mere misprints. Despite my objections, he disqualified nearly all of them. This unilateral retroactive change in the deal, which I would later encounter often on Wall Street, done by someone for their benefit just because they could get away with it, violated my sense of fair play. I quit reporting additional corrections.
Elliot Temple
Apr 28, 2018, 2:40:33 PM4/28/18
to FIGG, FI
On Apr 28, 2018, at 11:25 AM, Alisa Zinov'yevna Rosenbaum <petrogradp...@gmail.com> wrote:
> On Sat, Apr 28, 2018 at 1:36 PM, Elliot Temple wrote:
>
>>> On Apr 28, 2018, at 2:47 AM, Alisa Zinov'yevna Rosenbaum wrote:
>
>>> On Mon, Apr 2, 2018 at 4:10 PM, Elliot Temple wrote:
>
>>>> Another way to stand out would be to send money to everyone who has corrected you and to keep a list of corrections, with dates, and then people can see you often do listen, including recently, and genuinely value the help. And even if you rarely get successful corrections, you could still blog each unsuccessful correction you listened to and why you rejected it. People could then see how you handle input, criticize your handling of input, see that you do consider input in a reasonable way, learn from it, avoid repeating input you’ve heard before, etc.)
>>>
>>> That's a smart idea.
>>>
>>> I could offer to send money, say, to the first finder of any mistake in anything one publishes. I wonder what a good payment amount policy would be? I wouldn't want it to be large enough discourage me from publishing at all. On the other hand, I would want it to be large enough to incentivize people. Currently I'm still making typos, so it's enough to incentivize typical readers; I don't have to incentivize only FI experts.
>>
>> you could pay for first errors in blog posts only and then use a higher amount than you do with FI posts. if you write 5 posts per month on average, then $10/error (more than enough to get typos reported) would only risk $50/month.
>
> There are three good ideas in what you just said that I hadn't considered:
>
> - Paying only for the *first error* in a given piece of writing. I was thinking of paying for the *first report of any given error*.
What you wrote above is ambiguous and I read it as only paying for one error per piece of writing. So I didn't think I was telling you a new idea, heh. I didn't notice it was ambiguous, I just read it a different way than you meant it. Rereading now I see the ambiguity.
> On Sat, Apr 28, 2018 at 1:36 PM, Elliot Temple wrote:
>
>>> On Apr 28, 2018, at 2:47 AM, Alisa Zinov'yevna Rosenbaum wrote:
>
>>> On Mon, Apr 2, 2018 at 4:10 PM, Elliot Temple wrote:
>
>>>> Another way to stand out would be to send money to everyone who has corrected you and to keep a list of corrections, with dates, and then people can see you often do listen, including recently, and genuinely value the help. And even if you rarely get successful corrections, you could still blog each unsuccessful correction you listened to and why you rejected it. People could then see how you handle input, criticize your handling of input, see that you do consider input in a reasonable way, learn from it, avoid repeating input you’ve heard before, etc.)
>>>
>>> That's a smart idea.
>>>
>>> I could offer to send money, say, to the first finder of any mistake in anything one publishes. I wonder what a good payment amount policy would be? I wouldn't want it to be large enough discourage me from publishing at all. On the other hand, I would want it to be large enough to incentivize people. Currently I'm still making typos, so it's enough to incentivize typical readers; I don't have to incentivize only FI experts.
>>
>> you could pay for first errors in blog posts only and then use a higher amount than you do with FI posts. if you write 5 posts per month on average, then $10/error (more than enough to get typos reported) would only risk $50/month.
>
> There are three good ideas in what you just said that I hadn't considered:
>
> - Paying only for the *first error* in a given piece of writing. I was thinking of paying for the *first report of any given error*.
> Suppose I publish a piece of writing with 5 different errors. With the way I was thinking about it, I would risk having to make 5 payments. With your idea, I would only risk having to make 1 payment.
www.fallibleideas.com
Alisa Zinov'yevna Rosenbaum
Apr 28, 2018, 2:54:57 PM4/28/18
to fallibl...@googlegroups.com, FI
On Sat, Apr 28, 2018 at 11:40:28AM -0700, Elliot Temple wrote:
> On Apr 28, 2018, at 11:25 AM, Alisa Zinov'yevna Rosenbaum <petrogradp...@gmail.com> wrote:
>> On Sat, Apr 28, 2018 at 1:36 PM, Elliot Temple wrote:
>>> On Apr 28, 2018, at 2:47 AM, Alisa Zinov'yevna Rosenbaum wrote:
>>>> I could offer to send money, say, to the first finder of any mistake in anything one publishes.
Mismatch between "I" and "one".
> On Apr 28, 2018, at 11:25 AM, Alisa Zinov'yevna Rosenbaum <petrogradp...@gmail.com> wrote:
>> On Sat, Apr 28, 2018 at 1:36 PM, Elliot Temple wrote:
>>> On Apr 28, 2018, at 2:47 AM, Alisa Zinov'yevna Rosenbaum wrote:
>>>> I could offer to send money, say, to the first finder of any mistake in anything one publishes.
>>> you could pay for first errors in blog posts only and then use a higher amount than you do with FI posts. if you write 5 posts per month on average, then $10/error (more than enough to get typos reported) would only risk $50/month.
>>
>> There are three good ideas in what you just said that I hadn't considered:
>>
>> - Paying only for the *first error* in a given piece of writing. I was thinking of paying for the *first report of any given error*.
>
> What you wrote above is ambiguous and I read it as only paying for one error per piece of writing. So I didn't think I was telling you a new idea, heh. I didn't notice it was ambiguous, I just read it a different way than you meant it. Rereading now I see the ambiguity.
Your example of the $50/month risk rate is what made it clear to me that you were thinking of the bounty differently than I was. If you hadn't included that example, I would have assumed that you were thinking of the bounty in the same way as I was. This is an example of how examples can be helpful in communication.
Alisa Zinov'yevna Rosenbaum
Apr 29, 2018, 3:47:22 AM4/29/18
to fallibl...@googlegroups.com, FI
On Sat, Apr 28, 2018 at 5:47 AM, Alisa Zinov'yevna Rosenbaum wrote:
> I could offer to send money, say, to the first finder of any mistake in anything one publishes. I wonder what a good payment amount policy would be? I wouldn't want it to be large enough discourage me from publishing at all. On the other hand, I would want it to be large enough to incentivize people. Currently I'm still making typos, so it's enough to incentivize typical readers; I don't have to incentivize only FI experts.
At 22:56 in "Writing Philosophy Emails – April 2018 #3. Video 3" [WPE], Elliot Temple read the above paragraph and commented (lightly edited):
> I could offer to send money, say, to the first finder of any mistake in anything one publishes. I wonder what a good payment amount policy would be? I wouldn't want it to be large enough discourage me from publishing at all. On the other hand, I would want it to be large enough to incentivize people. Currently I'm still making typos, so it's enough to incentivize typical readers; I don't have to incentivize only FI experts.
>> When you say, "it's enough", it's hard to know what the "it" is, or what you mean here. This is a kind of vague, unclear statement. I think it means, "So it's good enough for me to incentivize only typical readers because they're able to point out mistakes, and I don't need a better person to get at least one mistake pointed out." It's something like that. It reads awkwardly.
Elliot is right: "it's enough" is an unclear reference. The alternative wording that Elliot suggested is clear.
[WPE] Elliot Temple (2018-04-28). "Writing Philosophy Emails – April 2018 #2. Video 3: Overreaching, unemployment writing, misc" https://gum.co/WMFTH
Reply all
Reply to author
Forward
0 new messages