Ask HN: How can I recognize when I don't know what I'm talking about? | Hacker News Reader

Ask HN: How can I recognize when I don't know what I'm talking about?

34 pointsspiffytech51 comments

Like most people on HN, I'm a pretty smart guy. Like most geeks and nerds, I have know a little bit about a lot of things, and a lot about a few things. There are plenty of topics I have minimal experience with, but have read enough about that I feel qualified to discuss, at least in a "contribute to the conversation" or "Basic Intro to Topic $FOO" kind of way.

The problem is I wind up in conversations where I think I know enough about a topic to talk about it, only to get called out by someone who knows I'm very wrong and can prove it. It's a normal part of life for this to happen sometimes, but I notice it happening too often as of late.

How can I recognize when I don't know what I'm talking about, before getting called out on it?

Edit: I'm humble enough to admit when I'm wrong, I just want to be wrong less often. Part of my problem is getting called out too often, but the other part is knowing when to keep quiet when I'm not around people who could correct me. I don't want to misinform my friends.

View on HN

There's a distinction that I've seen people miss, which I'll describe by example. Suppose you don't know Lisp, but you've seen many respectable people such as Paul Graham and Steve Yegge describe how great Lisp is for metaprogramming. This doesn't mean that you know the fact "Lisp => great for metaprogramming"! What you know is the fact "Many respectable people think => 'Lisp => great for metaprogramming'". This is an important distinction! It means you probably shouldn't talk about Lisp and metaprogramming, and if you do talk, you should make sure that you deliver the right fact, i.e. "I've heard that Lisp is good for metaprogramming", not "Lisp is good for metaprogramming".

> What I fear is looking incredible.

I'd rather that you fear being incredible (which it appears you are).

A resource that helped me think much more precisely about what I know and the level of confidence I have in that is the rationalist writings of Eliezer Yudkowsky. You might like to check out the sequences: http://wiki.lesswrong.com/wiki/Sequences

When what you know differs from what is real, you have a problem where your mental map does not match the actual territory. The most important sequence, Map And Territory, addresses that: http://wiki.lesswrong.com/wiki/Map_and_Territory_%28sequence...

I also recommend to you the Reductionism sequence: http://wiki.lesswrong.com/wiki/Reductionism_%28sequence%29

How often do you say "I don't know"? I think Voltaire was on to something when he wrote: "He must be very ignorant, for he answers every question he is asked."

> Suppose you don't know Lisp, but you've seen many respectable people such as Paul Graham and Steve Yegge describe how great Lisp is for metaprogramming. This doesn't mean that you know the fact "Lisp => great for metaprogramming"! What you know is the fact "Many respectable people think => 'Lisp => great for metaprogramming'".

An excellent distinction.

I think OP may be looking for http://wiki.lesswrong.com/wiki/Mysterious_Answers_to_Mysteri..., that's the one with e.g. http://lesswrong.com/lw/iq/guessing_the_teachers_password/

I've experienced the same problem. I like to think its mostly fixed now. Here are some things that worked for me:

Don't shy away from the ones that call you out, at least if they are giving you friendly pokes rather than being mean. The way I talk (in casual settings), I grew to like saying stuff like "What?!? Really?!? Oh!. I'm full of s--- then!" The value of these people is that they help you learn to recognize the internal cues of when you might be going off the rails. They can probably sense it in your voice tone and body language before you even get the bogus content fully out. You can learn to recognize that in yourself.

Do watch out if you're one of those (like many of us) who likes to talk a lot about stuff we "know" and likes to talk excitedly and at length. Watch some old cowboy movies that feature smart, strong silent-type heroes. If you don't have one already, buy yourself a cowboy hat as a personal reminder. (Substitute any similar character type for "cowboy" - your style choices are yours.) If you feel like you just have to say every clever thing that pops into your head in a conversation, work on not doing that - even for topics about which your knowledge is definitely good. (As a side effect, this makes conversations more interesting.)

Careful if, like many of us, you find it friendly and fun to tell people "Hah, no, you're wrong there! Let me tell you...." Among some geeks/nerds, that really is fun and playful. Among many people, it's not. It's also one of the leading causes of spouting bogosity, studies have shown.

Remember that its just human nature, nothing new. What was it Mark Twain said? Something like "It ain't what you don't know that gets you into trouble but what you know that just ain't so." Something like that, anyway. Close enough.

I use the advice of PG (which he observed from RTM) http://paulgraham.com/heroes.html

    Don't say anything unless you're fairly sure of it. If you're not omniscient, you just don't end up saying much.

    ... the trick is to pay careful attention to how you qualify what you say.

Start qualifying everything. Eventually you will stop commenting on subjects you don't know about when your qualification becomes: "well, I don't really know anything about this, but..."

The other aspect of this is to dig deeper into subjects. People think they know something about a topic because they read a couple of blog posts. Dig deeper- look at primary sources. Look at citations. If there aren't citations the statement is suspect, and you will qualify the knowledge as suspect when it comes into your brain, increasing your chances of qualifying when it comes out of your mouth.

> Start qualifying everything.

For some reason I find it annoying when people do that. For example I often hear something like ("I don't know but, the library you are using is bloated compared to library X").

I frequently catch myself saying "I don't know" in front of assertions I make. This is an issue when talking to a less knowledgeable person, because the "I don't know" puts the more knowledgeable person at an immediate disadvantage, and lets the person with less knowledge have more of a say in how things should be run.

I find better just to confidently state something, and get called out for inaccuracies (it's certainly more memorable). Also think about what you said at the end of the day. It's better to recognize and admit inaccuracies than to hang onto your pride (http://www.google.com/search?sourceid=chrome&ie=UTF-8...).

> I find better just to confidently state something, and get called out for inaccuracies (it's certainly more memorable).

Of course this is dangerous when talking to people who know less than you about something, since they may not have the ability to call you out.

To that, if you do see something wrong with something you said in the past, it is morally imperative that you correct the mistake.

Try turning the statements into genuine questions. For example: "I don't know but, the library you are using is bloated compared to library X". That could be turned into "Isn't library X much smaller than the library you're using? Do you need that big of a library?" Or even "I thought the library you're using is considered bloated compared to library X." It's a gentle way to get the other person to explain themselves.

Those are good ways of bringing up the point.

"I don't know..." is not a real qualification. Qualifying things also means working at doing proper qualifications. "I read a comparison of library X and Y a while ago that stated (qualification) library X is bloated to library Y, so we should investigate that before making a decision as too which one to use"

It is a great point that people don't qualify because it makes them sound less confident, and people that sound more confident end up being more convincing to most people. So there is a cultural problem there (hopefully confidence does not sway a scientific group).

Cultural problems are tough- to maintain your stance you have to try to change the culture (sometimes impossible) or move to a different culture. But you can change the culture a little by asking "where did you read that?", or "so you have used/did that yourself?". Obviously you have to do this in a way that doesn't annoy.

You have to work at finding the middle road between qualifying and giving an impression of relative confidence.

Let me qualify this by saying these are just what comes to my mind on this issue. I would very much appreciate the pragmatic advice of others on this concept.

> So there is a cultural problem there (hopefully confidence does not sway a scientific group)

Isn't this the norm though? Given all things equal, it's been my experience that people who make quick and strong judgement tend to have more control over a social situation.

Of course somebody who is wrong often will have less reputation. However, being wrong is not always bad, especially if it leads to further thought and exploration to find the truth and the person is willing to change one proven wrong.

> It is a great point that people don't qualify because it makes them sound less confident, and people that sound more confident end up being more convincing to most people.

Your qualification about reading the comparison instill confidence on the rest of the statement. Maybe it's the self-deprecating qualifications that should be avoided...

> You have to work at finding the middle road between qualifying and giving an impression of relative confidence.

True.

---

Some meta thoughts. I would have to expend lots of energy and time to be able to properly qualify some of my statements. I find the time/accuracy trade-off to skew toward time, here, with the hopes that a well educated and astute crowd (such as HN commenters), can point any flaws or provide validation to my statements.

This is not that big of a problem and I think to varying degrees, everyone does it. Learn to spot when you do it more frequently. And for practice, make a note when others do it as well. You'll learn to distinguish air-tight arguments from sloppy ones. It also helps to use a mental filter to call yourself out before anyone else can. Here are things that help me:

1/. State an argument. Then, take the opposite side - ask yourself, what are the arguments people might have against my argument? What assumptions am I making? Is what I am saying true in all cases or only under specific circumstances?

2/. Reading Richard Feynman's Cargo Cult Science speech at Caltech helped: http://www.lhup.edu/~DSIMANEK/cargocul.htm

3/. Two quotes from Charlie Munger (Warren Buffett's partner), that also act like a filter:

"You are not entitled to an opinion unless you can state the arguments against your opinion better than your opponents can."

and from an interview:

Q: You've often said that one of the keys to your success has simply been to avoid making the garden-variety mistakes that you see other people make.

Munger: Warren and I have skills that could easily be taught to other people. One skill is knowing the edge of your own competency. It's not a competency if you don't know the edge of it. And Warren and I are better at tuning out the standard stupidities. We've left a lot of more talented and diligent people in the dust, just by working hard at eliminating standard error.

First, in case you're wondering if you should just keep your trap shut, it's ok to get called out once in a while. It can be embarrassing, but it's part of every learning curve & it will make you smarter, so long as you listen and handle it with grace. There is a lot to know in the world, and you won't be an expert in much, but it's good to share the knowledge you have. It's good for novice observers to see you called out and corrected, and for them to see you reply with appropriate solicitousness.

Second, it's sometimes really hard to know. The best advice is to avoid speaking with absolute authority, unless you have evidence in hand to back up your shit. Even the experts get tripped up occasionally.

I guess my advice boils down to this: embrace being corrected because it means your knowledge has just been expanded, and approach all topics with a modicum of humility.

I love being corrected, because it's educational. What I fear is looking incredible. Get corrected too often and people start to assume whatever you say is probably wrong

I think you focus on the wrong problem. It is ok to be wrong and if someone can prove it to you, be happy to have learn something new and thanks him/her.

Of course, to do that, you need to have an open mind and always act on the safe side. Instead of saying: "You stupid idiot, haskell is not purely functional because there are monads", you're better with "I'm pretty sure haskell..." or "I think Haskell..". This way, you encourage people that might know more than you in that subject to speak and help you instead of aggressively attacking them.

There's nothing wrong in being wrong. And to be honest, I think it's the only way to learn :o

I have the same problem, and so does my dad. My safety catch is to ask myself if I could begin working on a project related to what I'm talking about. If not, then I know I'm full of shit. It doesn't mean you have to stop talking; you just have to let other people know that you're full of shit, and simply making educated guesses.

Do you have a conception of the difference between stuff you "know" (i.e. certain facts that you are sure are correct) and things you "believe to be the case"?

If yes, are the things you get called out on stuff you had previously categorized as stuff you "know"? If so, you need to think on the distinction between the two.

If not, then, what I do is this: If I know something to be correct I am happy to assert it. If I only believe it to be the case, I will caveat it with something like "It is my current understanding that..." and make it clear that I am not the font of all knowledge on the subject, and others may have a more informed opinion.

You can mentally prefix this whole answer with a "It is my current understanding that.."

I'm having this exact same problem with a colleague at the moment and my root cause analysis is that he's not making the right judgement about the knowledge level of the other conversation participants.

I would suggest trying to a) listen and b) gauge the level of experience of the other guys in the conversation. If they have 'doing' knowledge that automatically trumps your 'I read on the internet' knowledge so, yknow, silence is golden - everyone has access to google...

If you have the context to sit a level above the conversation and can see that the issues being talked about are clearly peripheral, you know enough to redirect the conversation to the core issues with a good question.

If you don't, you can ask clarifying questions until you do. Always focus on what would help you decide one way or another.

If you get into a disagreement with someone and you can't resolve it in the scope of the conversation, ask for a reference or data.

If you're in a conversation with an expert, you're the student. And that's ok. We're all students of the world.

At the same time, don't be afraid of speaking too far ahead of your knowledge. When someone calls you on it, you have a chance to learn.

Become a questioner.

The real problem is if you've been fronting and cracks appear in the facade. Then people know beyond the shadow of a doubt that you've been acting small and think less of you. The only cure is not to be a bullshitter. Don't pretend you know what you don't.

There's a difference between pretending like you know it all and being forced to act on imperfect information. The former is a character flaw, the latter is just life happening.

I think your friends may have just pegged you for a BSer and now anytime you say something that doesn't sound right they are going to call you out on it, even if they don't necessarily have the facts to back themselves up.

I like to infer. Read a little bit and then infer a lot from it. To reduce being called out on something, explain how to jump from point A to point B. This actually serves 2 purposes.

1) Reduce the chances of someone jumping on you because they think you are a BSer based on your history of BSing.

2) Enhances your thought process because you become aware of errors in your thought process when you are called out.

Ultimately #2 is most important. When you are called out or corrected, you gain insight not simply in your own thinking process, but how the person in your conversation thinks as well. Listening to your criticisms and implementing them into your thought process to correct for them in the future makes you seem way more credible and increases your ability to infer in the future.

You could try to only have contentious opinions on topics that you are sure you are right about. On other topics take a listening role until you've heard differing opinions. You could of course offer an opinion you may have as a question:

What do you think of the opinion many people have..which is Do cats always land on their feet...even when they've been buttered e.t.c.

Opinions are one thing, having your facts wrong is another.

The greatest thing about you is that at least you admit it.

As a nobody I think that the biggest sign people are venturing into areas where they don't know what the hell they are talking about is when they start using buzz words, technical terms and oft quoted stuff a lot. I think that when people have thought a lot about something, or truly understood it they tend to talk about it in their own words, and they don't talk about terms or something arcane but concepts behind them. They take into context things that just aren't at the surface, and then base their conversation on that.

This isn't a hard and fast rule though. As it varies with a lot of people, but it is really useful for me as I have to turn to 3rd party sources to learn stuff pretty often (a textbook or a teacher is a first and second part resource. You know if you can trust them or not. Online help sources whereas are another game altogether).

It's good not to adopt strong opinions on things you don't know much about. There are also lots of ways to drop in disclaimers about your (lack of) knowledge so you can still have a good conversation. This results in more learning and less arguing.

You can't, all you can do is be open and humble when you're wrong. When you say something that turns out to be incorrect, don't get defensive or extremely, apologetic - a simple "Oh, really. Interesting" will suffice and move on. It can be difficult if the other person is extremely insecure and is trying to shove it in your face that you were wrong, but so long as you yourself are secure then you will be ok with being wrong. The smartest people I know are also the most humble because they are open to new ideas and willing to be wrong.

Yes, you can. Especially if you're not an expert, sometimes talking a bit less, and listening more, is certainly a start to fixing this problem.

Try to refute yourself mentally before you speak. If you can argue on both sides of a topic, then you aren't really sure.

The best way to proceed when you're not sure is to ask the others. Don't say, "The internet is made of turtles," instead say, "hey, I heard it's turtles all the way down, do you know if that's true?" And listen. That's how you fill in those gaps in your knowledge.

I live by the axiom, "It is better to remain silent and be thought a fool, than to open your mouth and remove any doubt."

Every time you feel tempted to jump into a conversation with an assertion or statement, first rephrase it as a question, then try to answer it.

That will give you an idea of whether you really know what you're talking about. Then decide from there whether you can submit it as an assertion (yes you do know what you're talking about) or as a question (maybe not as you thought).

It's simple enough to fix. Just stop making assertions and ask questions instead. Not only will you no longer have to worry about foot-in-mouth disease, you will also be more liked! And yes, I know I just violated my own advice.

When words are many, transgression is inevitable. But he who keeps his peace is wise.

Wise men speak because they have something to say; Fools because they have to say something.

Presence.

Zen.

What? You're phrasing something that you don't actually know as a fact that you know? That's called lying. You recognize it by learning to not lie and deceive people as a child.

Not sure what else to tell you. If you're passing off knowledge you pulled out your ass and are getting called on it, maybe you shouldn't do that. If it's speculation or if you're trying to be helpful but don't know for sure, then phrase it that way. Or if you can't take the criticism just keep your nose out of it.

I don't understand where the disconnect is to be honest.

Speaking a mistruth and lying are not the same thing. I sometimes think I'm telling the truth, but my facts are wrong. I'm not trying to deceive.

I'm not just pulling things out of my butt, either. For example, I spend a lot of time researching new technologies, but rarely have the time to become proficient with them. This is very similar to the kind of learning you may do in a classroom; I've never made an electromagnet, but thanks to my physics class I can tell you about them.

Sometimes I say something incorrect. Sometimes someone with more experience disputes my claims and I don't have the knowledge and experience to rebut them, or to know if they're really right.

Modern society is founded on feeding off of each other's knowledge and experiences; if we all spoke only about what we knew very well, and first-hand, we'd say all very little. I'm humble enough to admit when I'm wrong, but I'm trying to be wrong less often. It makes me look incredible.

I guess I still don't understand what you're looking for as an answer. Logically there are only a few paths to "be wrong less often".

1. Become an expert in everything.

2. Stop talking about the things that you're unsure of.

3. Preface the things you're unsure of with "I'm unsure, but I think".

If you're worried about looking incredible (and not in the good connotation) because you are saying too many things along the lines of "I think..." then stop doing it or learn to be confident that your expertise in the places where you have it will shine.

What others paths are there here for "be[ing] wrong less often"? Am I missing some hidden deeper question at play here?

I do preface things I'm unsure of as you suggest, but sometimes I'm wrong about things I am sure of. I'm trying to figure out whether this is a problem with my definition of "sure of", whether I just need to do more research and get more experience on topics in general before talking about them, or whether there's some indication I may not know what I'm talking about that I just haven't found yet.

I'm asking in the hopes that other people have solved this problem for themselves, and could say how.

If you were misinformed or were just simply plain wrong (aka you thought you knew something as solid fact) then... you were wrong. Life goes on. I mean, if this is happening a lot, you properly need to question the threshold at which you consider yourself having "known" something, or you need to read more closely, but being wrong isn't the end of the world.

> I'm trying to figure out whether this is a problem with my definition of "sure of"

I should this is obvious. If you're "sure of" false things a lot, obviously something is going wrong. Try not to be so sure of things.

I generally try to source everything I say, if it's beyond a certain amount of common sense or complexity. Just the act of looking up sources before hitting that button (and often in the middle of writing the post) help me catch most of my mistakes. At least, I haven't been called out on a mistake very often. If I can't source something, I usually clearly note that in some way.

I get the feeling he's more concerned about in person conversations, where looking things up before you speak is not really feasible (yes, you have your iPhone, but seriously...)

4. Get better at generalising knowledge and reasoning to new areas.

When you get it wrong, look at why. Were you missing a crucial fact? Then how could you have found that fact out, and other facts like it. Is your reasoning faulty? Get better at reasoning. Did you make an unfounded assumption? Check your assumptions better. Did you ignore something without good reason? Don't. Is their something special about working in this field? Learn that stuff. Are you expected to have background knowledge? Grab some of that.

... I guess that's really a solution to 1 though.

Calling it lying is going overboard. Sometime what happens is that people overgeneralize from the domain they know to domains they don't. Engineers, because we are problem solvers, are especially prone to do this, whenever we are given a problem, we like to offer a solution.

I feel like I'm repeating myself in my replies and I do apologize, but I feel that there is a distinction between outright making something up, asserting assumptions as false and being wrong. I think that if you preface something with "I think...", you're disclosing the fact that you are unsure or are making an assumption and that that falls outside the realm of this discussion, generally.

Disclosure works in the situations when one is unsure of one's facts. The question extends beyond that situation, though. When you say something wrong that you were sure of your credibility is damaged at least as badly as if you said something you were unsure of.

I appreciate the advice of disclosure that you and many others in this topic have given, and I will definitely be more arduous in disclosing when I am unsure of something, or just plain stay quiet. I would, however, still like to address not speaking incorrectly when I'm sure of my facts.

It's not lying. For some people "fact you know" is not a well-defined category. If you are smart and good at inferring stuff, and you know some facts, then maybe you can go further. Maybe you can provide insights. You can certainly appear more knowledgeable than you are (you aren't, you're working things out as you go along, which isn't guaranteed to be correct.)

What it is unfortunately is intellectually arrogant. I saw a great comment from roel_v (http://news.ycombinator.com/item?id=1535064) about this (reproduced below). If the benefits are working out for you (you look smart, you're right often enough) then great. But if you keep getting called out, and you're worried about that, then you need to equivocate more. You also need to become more aware of the unknown unknowns. It's all very well knowing some things, and guessing others, but you also need to know what you actually know and what you don't, otherwise you'll not only look stupid, but at some point you'll also do something really dumb too.

"95% of legal chatter amongst technical people on the internet is meaningless to lawyers ('meaningless' as in 'jibberish', 'flux capacitor'-style nonsense) - because the fundamental assumptions and reasoning methods are so vastly different"

You don't understand, yet feel qualified to attribute his problem to a character flaw.

Interesting. It appears you may have the same problem as the original poster.

I didn't mean to suggest that he is a compulsive liar or is a bad person, but I think it's deceitful to pass assumptions off as facts. I don't know whether he is specifically referring to making things up (which I see as deceitful and thus lying), asserting assumptions as truth (which I see is a gray light), or simply being wrong. Being wrong is one thing. If I tell you 2+2=5 and you tell someone else thinking that it's a fact, then you aren't acting intentionally in your deceit. If you just start making assumptions on your own and assert them as truth, I feel like that takes a certain amount of effort in a dishonest direction.

By your own logic you are a liar. Fortunately most dictionaries disagree with you.

http://dictionary.reference.com/browse/lie http://definr.com/lying

To safely call a statement a lie according to all definitions it must be false, but more importantly it must be said knowing that is false (having an intent to deceive).

I feel like I differentiated between making stuff up, being unsure and simply being wrong. The 'lying' is mostly an attribution to making things up or asserting assumptions as fact which I still would consider deceitful.

edit: I may not have differentiated that as well as I had thought in my original post. In another comment on this comment level, I tried to differentiate between the various types of "being wrong" and/or their causes.

Actually one of the most annoying aspects of HN is the people who comment on a subject as if they were "laying down the law", even though they're wrong. Speaking with a voice of authority when you don't actually really know that what you're saying is true could be called "spewing bullshit".

I don't see being proven wrong as a very big deal. Say the wrong things, be proven wrong, and now you know.

Way easier than to second guess yourself every turn.

In startups we talk about failure as an important part of the learning experience, so it applies to life in general too.

I have to partially disagree. While I definitely don't think you should second guess your every turn, it's probably good not to go repeating things (especially unusual things) until you've at least tried to verify them. That's where urban legends come from, and its part of why its so hard to actually get the truth out. It gets lost in so much noise.

Reply on news.ycombinator.com