Hacker News new | past | comments | ask | show | jobs | submit login
Assume positive intent (rickmanelius.com)
237 points by mooreds on Aug 17, 2017 | hide | past | web | favorite | 131 comments



1) It’s 100% reasonable to have a high degree of skepticism within a low-trust environment. For example, I would never assume positive intent and allow my daughter to be alone with a registered sex offender just because the person claimed they had changed. I would also never trust an alcoholic with a house full of liquor. Once a person has violated trust against a particular metric, it's OK to take a different position in order not to put yourself in harm's way.

You can only assume positive intent in situations where intent is not clear and you need to infer or assume. Once there is a proven track record of bad acts clearly establishing intent, assuming positive intent amounts to deluding yourself.

If you don't know, then assuming positive intent and looking to situational factors as the source of problems can be surprisingly powerful. But, once you have affirmative evidence of bad faith, then you need to go with that. Assumptions about motive only make sense under circumstances where motive or intent is unclear.


So then we have to get very good at determining if something is unclear.

This is where most people make the most mistakes, they assume their own judgement is sound. Hence the article.


A bit of a nitpick, but

>I would never assume positive intent and allow my daughter to be alone with a registered sex offender

I don't know, prehaps that's true in the case where one doesn't know what the sex offender has done (i.e what crime got them onto that list), but it's not sufficient in general.

Although most people on the list are there for good reasons, there are not-good reasons too, such as urinating in a public park at night, having a lewd picture of your 17 year old girlfriend/boyfriend on your phone, and in some countries possessing/writing certain pornographic comics or stories.

So I think it is too dismissive, similar to how it is also dimissive to say the same thing about an ex-convict.


Substitute "known child molester" for "registered sex offender" and see how you feel about that statement. I am sure he also would not want to leave his child with a child molester who was not yet known to be a child molester, but most people have no idea how to effectively screen out such people. So, they use metrics like registered sex offender.

While I agree that most parents use crappy rubrics for deciding whom to entrust their kids with, I am not really comfortable with arguing in favor of trying to do some kind of justice for some edge case adult over a parent looking out for the welfare of their own child. The parent has a duty to their child. They do not have a similar duty to some guy who took a leak in the park, yet their dream job is babysitting. Depriving public urinators of some right to become a babysitter seems a rather small thing in the name of protecting our children from possible child rape.

The place to fight this battle for justice is with the system that designates public urinators et al as sex offenders, not with parents trying to decide whom to leave their kids with.


Hanlon's Razor (‘Never attribute to malice that which can be adequately explained by neglect.’) is better ... just assume people are not trying to screw you right away.

The lack assumed negative intent is a positive in my book

https://www.farnamstreetblog.com/2017/04/mental-model-hanlon...


That does seem better, because I don't think I would say people generally have positive intent. It's more like benign or neutral intent.


The way I like to put it is: People aren't for you or against you, they're for themselves.

Most of the time people don't intentionally slight you, they just don't care enough to actively avoid it.


> People aren't for you or against you, they're for themselves.

Where does bullying fall in your classification?


One popular theory of bullying states that bullies are uncertain of their own position in the social hierarchy, so they attempt to bolster it by pushing down on those weaker than them.

http://www.paulgraham.com/nerds.html


Which means that to be "for themselves" they feel the need to be "against others" as they view things as zero sum. I guess similar applies to other zero sum situations (or situations that people view as zero sum). The car salesman isn't trying to screw you, he's just trying to make more money for himself, by screwing you.


I think its safe to assume in general people are just trying to do the best they can, whether they have the tools and support they need to get there is another matter


I do not agree. Many people just do enough to get by or say get promoted. I would hazard saying most.

It is a rare person that really wants to do best and they are typically not rewarded for it. Especially when they don't get rewarded.


Unless you are trained by high-performing sociopaths (e.g. by having them in your own family) and then seeing half-baked "mastery" of those techniques in almost everyone you meet, as they are subconsciously aping those "masters" that became successful. Once you start seeing this, it can't be unseen; good luck not feeling disgusted all day long.


There is one thing that people are binary about: Their feelings for sociopaths. You can immediately tell who has had to spend time around an honest-to-goodness sociopath and who hasn't. I always say that sociopaths are insects in human skin. Once you've had a brush with one of these "people", it forever alters your perception of the world and the people that inhabit it.


I can tell you from personal experience that having faulty empathy wiring really sucks, and people who suffer from this problem are deserving of sympathy, even if it's at arms-length. It shouldn't be a surprise that going through life feeling cold and isolated is going to result in someone who does shitty things.


Which techniques?


[Gaslighting](https://en.wikipedia.org/wiki/Gaslighting) is perhaps one such example.


bitL suggested that you are better off not knowing. Are you sure you want to go down the avenue bitL explicitly suggests will disgust you all day long?


Came here to say this.

This kind of work irks me. No offense to the author, this is a widespread trend, not a single example.

You write a whole blog-post on a topic and not mention Hanlon's razor. It is inexcusable that you didn't know about it. It was your job to do "literature survey" or some google search, or an intellectual discussion with a handful of people, (ideally all three of them, and more) before you blurt out a whole essay on a topic. And then in the essay you create an image of yourself originally coming up with such a brilliant idea.

This is just a piece of writing (and maybe I am a bit harsh but I'm just trying to point to a trend). We're on HN. This is way worse when it comes to coding/github projects:

You create some kind of code, put it on github, create a blog-post in some shiny flat/material design, grey font, maybe medium or nautil.us posting. And put that posting on tech forums, HN, linkedin, reddit, whatever.

And you fail to mention other projects preceding you on this same exact topic, much less discuss the pros and cons of your approach vs previous approach(es). Who knows you might've ended up finding out your own code was a complete reinventing of the wheel and completely redundant (maybe you're too afraid to find that out). (As an example, look at my recent comment about 'building your own linux' vs LFS [1]).

Creating something new and original is hard work. Part of that hard work is searching for existing efforts, related work, doing a literature survey, and whether or not they were successful, and how they compare/contrast with what you did or are going to do, and so on. Short of that, you're just contributing to the noise, the bloat, and the information overload of your content consumers.

What's worse. Not putting in the full effort works for these people too. More often than not, such people are looking for a quick karma-whoring to meet their immediate objective. A github coder is looking to put something quick on their resume, to land a job interview. It's the job of the interviewer or potential employer to hold such people accountable, but I doubt they do.

I don't know. This irks me to no extent. I'm big on minimalism. I spend a lot of my efforts making sure I don't say something that's already said ('say' = a piece or writing, or a piece of code). I don't like to contribute to information bloat. Maybe it's a self-defeating habit (i.e., if you don't say anything most of time, people think you have no talent, you're a nobody).

edit: It also reminds of NIH (not-invented-here) syndrome, something Google is known for. To be clear, it's worse than when it's done by noobs (e.g., for quick karma-whoring to land a job interview, at least they're desperate). When you're in a position of "status", you definitely can do better. It also reminds of the website LessWrong [2]. Those self-entitled wannabes have created a whole universe of philosophical/logical terminology for themselves without consulting the outside world and trying to find out whether "their ideas" originated elsewhere or not.

edit 2: In case anyone wondering, yes I am advocating the opposite of 'publish or perish'. And 'publish or perish' is part of the reason we have so much information bloat (on the internet, in the libraries, everywhere). And that's part of the reason why I say my approach is self-defeating.

[1] https://news.ycombinator.com/item?id=15030582

[2] http://lesswrong.com/


Once you've earned a bit of trust with the approach in the article the above post shows you how effectively to destroy any relationship you've now built.


> You write a whole blog-post on a topic...It was your job to do "literature survey"

That's a pretty high bar to impose on a guy's personal blog site. I mean, he also has a poetry section. From clicking around on his blog it also appears he has a PhD from MIT, so he's certainly no stranger to research, if that were his project in this post.

> And put that posting on tech forums, HN, linkedin, reddit, whatever.

It doesn't look like that happened here. The blog was written by a guy named Rick Manelius, and posted to HN a guy named by Dan Moore. While it's possible there's some collusion going on here, I say we "assume positive intent"; that is, assume Dan found Rick's blog entry insightful and posted it to HN for the rest of us to consider.


I see where you're coming from. As I get older I keep seeing the same things reinvented, only with all-new problems (or even the same problems). That said, I invite you to consider two things:

1) Since this is going to happen regardless of how you feel about it, arguing that human beings "should" do something differently just leads to dissatisfaction on your part. I simply roll "cites prior work" and "compares to similar approaches" into my set of criteria for judging the utility of a work or idea.

2) There is inherent value for learners in revisiting familiar (to us) concepts on their own terms. If they get excited about learning something or producing something that has been around for decades, great. They don't diminish the other works by producing theirs. And sometimes there are kernels of something new.

That said, I do require every engineer who wants to introduce some new piece of technology to do a "literature survey" as a means of teaching them how to think critically about problems.


That's a very low bar for an essay... at least in terms of length. I'd be surprised if this took long to write which isn't a criticism so much an expression of confusion as to why you called it an essay. It's also a little weird to call it inexcusable they didn't acknowledge Hanlon's Razor because it's hardly a common reference. Even the wikipedia page only gives two instances of Hanlon's razor being used and one of them is an acknowledgement that the general idea is very likely much older than the Jargon File glossary entry.


As I said, I admit I might be a bit harsh. I took a slightly bad example of what I had in mind and went ahead and said what I wanted to say.

The bulk of my comment is about the general trend, as well as the coding community.


I feel you, I do this so often :/


This reminds me of another comment I once read...


I love this comment. Maybe because I am a software engineer and dig the comments on the lack of precision and research. But also because I'm a manager and had to talk many developers of a cliff who got irked when things are not right or complete.

Applying Hanlon's Razor, I don't see any malice in this comment at all. It's just a sharp comment with a logically valid point that is a little off base socially. Behavior like this is a feature of analytic/software/STEMy people, not a bug!


I've always held the belief that"reinventing the wheel" in this context is at worst a neutral thing. At best it is part of iterative progress. Perhaps the wheel itself is the same, but one small part could be slightly better. Sharing your findings is a part of this.

What really seems to grind your gears is the curation. The fact that this "inexcusable" blog post made it to the HN front page. Blaming the author for not living up to your (extremely ambitious) standards seem silly to me.


Given that almost every original thought as we know it is a synthesis of many different pieces of data rather than a burst of light out of a void, your argument is based on some very liquid ground.


What's wrong with recreating and repackaging ideas?

You'll gain a better understanding of the ideas, show others how to do the same (if you publish your efforts) and perhaps create new ideas through accidental new combinations or the concerted effort to integrate multiple existing ideas in an unusual way.


> And you fail to mention other projects preceding you on this same exact topic, much less discuss the pros and cons of your approach vs previous approach(es).

I think the issue isn't the recreation/repackaging but the omission of an honest discussion of how the new approach differs or what is done differently.


You're speaking to an important point and I often share this frustration. I don't have much to add in a comment, but thought I'd say as much because the other replies are disagreeing.


Dude, this is so negative, what gives?


Hanlon is in the same-but-different territory, because it's possible (or even somewhat assumed) that Hanlon's incompetence comes with good intentions (since you've rejected malice).

Focussing on the good intentions instead of the incompetence may be beneficial for your own mental health.


It sure is interesting whether it's just good for your mental health or also good for your longevity and lifetime influence.


unless you are doing business in China, or with others with similar geo-political leanings. Oh wait, then how will you know?

This is why in high stakes environments this "assume positive intent" falls apart. Only where the costs don't really matter (and for a lot of wealthy americans, a couple hundred bucks lost on a bad contractor is low stakes) can you really afford to be so lax.


I used to think this way, and then I built a house. Try keeping an open mind about that process sometime and see how many supposedly top-notch and reputable contractors fess up to mistakes they've made. By my count, the numbers are reversed - maybe 1 in 25 will come forward and say, "I damaged this and I'm going to have to buy a replacement part. Sorry about that." If you discover a problem and then call them on it, I'd give it 50-50 odds that they will claim innocence and balk at correcting the issue.

It would not have been my first instinct to put professional tradesmen in some sort of low trust category, but that has been my experience, and I don't think I'm alone.

Are they being malicious? No, I don't think so. Just lazy and/or careless - but the aftermath is largely the same.


Not to disagree, but once the price on the contract is set and a mistake is either covered by you or them, I think you've entered the realm of a "low-trust environment" which is addressed by the author.


I think that just furthers my argument - mistakes get made all the time since we're all humans around here as far as I know. I honestly don't have any problem with people ( tradesmen or otherwise ) making mistakes as long as they own up to them and/or fix them. To say that "once a mistake has been made and a solution negotiated, we're now in a low-trust environment" flies in the face of "Always approach people with optimism".


> but once the price on the contract is set and a mistake is either covered by you or them,

In my experience dealing with tradesmen, it's very rare to get a full price up front.


The author discusses some caveats to this approach. We might also take a critical lens toward trust as a default stance by applying an understanding of systems of oppression. I love the way this is discussed in AORTA's facilitation guide (http://aorta.coop/portfolio_page/anti-oppressive-facilitatio...):

> There are a few community agreements that participants often bring up that we don’t tend to use or bring with us. Two of the most common ones are “assume best intentions” and “default to trust.” The reason we don’t use these is because when someone is unable to do this (say they’re feeling untrusting of someone, or unsafe), having a community agreement telling them to do so isn’t going to change anything. These agreements aren’t always realistic, especially when we take into consideration that when people have been harmed by sexism, racism, homophobia, transphobia, classism, they/we build up necessary tools to care for and protect themselves/ourselves. Agreements we offer instead that capture the spirit of these are “we can’t be articulate all the time,” “be generous with each other,” or “this is a space for learning.”


I love Linus Torvalds quote on trust (paraphrase, not exact citation):

"People can trust me because they don't have to."

He was talking in the context of Linux kernel, but I think this applies universally. If you are in a relationship where the other side has more power, you cannot trust them. However, if you have equal (or greater) power, and you can walk away, you can trust them.

And again, people intuitively know this, I think, so if you want somebody to trust you, then you should give them enough power over yourself, expose yourself a bit.


The right to be sued is all about this. People often think that the right to sue is more important than the right to be sued, but it's really the latter that enables commerce. Without it, nobody would do business with a stranger, because they have no way of knowing whether the stranger will uphold their promises, and the stranger has no incentive to. But by granting a neutral third-party the ability to enforce consequences upon you, firms can create trust, and with trust they can promote trade.

Actually, I wonder if a major part of the decline of trust in the U.S. today is because the price of litigation has spiraled outside of the means of most Americans. So if the cable company or Internet company or bank or debt collector screws you, there's little you can do about it, and so people assume that the system must be corrupt.


I find your comment extremely insightful, especially the second paragraph. So thanks for that.

It makes me wonder if we could bring back more "trust in the system" by putting a cap on the price of litigation, or using "public prosecutors" the same way we use "public defenders".

My understanding of the system is that poor folks only get "public defenders" when they are arrested and can't afford a lawyer, but apart from the government bringing cases against private individuals (eg, "the people vs larry flint"), we don't see poor private individuals bringing cases against the government using "public prosecutors" when they can't afford an attorney. Instead, we see things like the EFF, the NAACP, and the ACLU, who all have to raise large sums of money to pay for private (as far as I'm aware) attorneys to bring cases to court.

It seems that the bias in the system is like a grain of wood that runs from rich to poor, where poor is against the grain.

So yeah, I do assume that the system is a corrupt duck, because it looks, walks and talks like a corrupt duck.


> It makes me wonder if we could bring back more "trust in the system" by putting a cap on the price of litigation, or using "public prosecutors" the same way we use "public defenders".

Could an alternative be to evenly split the resources for both sides? So if a court case is brought between a poor person and a rich person (or poor company/rich company), the two total the amount they'll spend on lawyers and then get exactly 50% each to use for legal costs.

So if you have to say, sue the cable company, the latter in turn has to pay for a large percentage of your legal fees regardless of if they win or lose.

The court system might change a lot if both sides were guaranteed equally good legal teams...


The English rule on costs in the County Courts of England and Wales is codified in the Civil Procedure Rules part 44 [1] (although there's more in CPR part 36, and there's recent covering case-law. The costs rules are similar in the High Court and appellate courts: the general practice is "the costs follow the event" (i.e., the winner's costs are generally at least partly paid by the loser) so long as that does not offend justice or proportionality.

A "technically you won" judgment where the winner's costs were extremely large could offend justice or proportionality enough that there is an atypical costs order favouring the winner (e.g. the winner gets at most only a small amount of its costs paid by the loser).

Costs tend to become the elephant in the room during litigation, since they increase in dramatic steps over the various phases. This tends to pressure parties into compromising their positions and settle on costs out-of-court on some privately agreed basis, rather than risk an adverse costs order if the opposing party is slightly successful. If, for example, a defendant makes a counter-claim that offsets (say) half of the claim, or simply an offer to settle for half of the claim, and the claimant insists on going to a full trial, the claimant risks having to pay both sides' costs if the defendant is successful in the offsetting counter-claim or if the defendant is only found liable for half the claim.

Cost orders often state that costs are "to be assessed if not agreed"; if the party who will be paid costs under the order has an unreasonable accounting of its costs, there is an assessment procedure wherein the court can strike out or reduce some of the costed items. Often where a very expensive lawyer is doing a lot of work compared to the price of the lawyer (and the amount of the work) being done by the party paying the costs, the courts will tend to reduce the amount of costs payable, on proportionality grounds.

Finally, prevailing parties do not fully recover the full amount they spent on the litigation in many circumstances.

The English Rule tends to contain costs, although it does not always do so. Moreover, it has the perverse effect that meritorious cases that would establish new law rarely make it all the way through trial and appeal, because when a party feels its position is less than fully solid and is seeing costs mount on both sides, it will generally try very hard to settle. A party that rejects these attempts to settle is also at risks with respect to orders on costs, so will generally accept. As a result, there is no final judgment, and the law remains unchanged no matter what the parties agree in their settlement; these cases aren't even generally reported. Unfortunately the legislature is not very quick to tidy up areas of private law where this is most common (the English law of real property is the most extreme case of this) and many cases that would help resolve oddities in the law simply never go far enough before the courts before settlement, so the oddities remain.

Additionally, under the English rule, withdrawing from a case because one does not have the money or energy to continue it to its conclusion generally has costs consequences for the withdrawing party: they pay the costs of the remaining party more or less as if they had lost. An extremely deep-pocketed party facing a party with shaky finances (even if the latter has a decent case) is often tempted to rack up significant costs and use the threat of an adverse costs order to "force" a withdrawal from the case early on. So the English rule is not really a cure-all, unless the smaller party is able to convince a judge to make an order that goes to the heart of proportionality, and that is extremely unlikely in the early phases of litigation. (On the other hand, many judges are getting fed up with the amount of costs racked up by parties during the disclosure phase, and there is recent case law that is likely to rein in this sort of abusive behaviour by deep-pocketed parties, even when they face-off against other deep-pocketed parties.)

Generally, giving discretion to courts to award parties costs (or not) based on principles like justice, proportionality, and (un)professional behaviour during the litigation, is seen as a decent step towards resolving some of the access-to-justice problems that strongly advantage the most well-funded parties at the costs of everyone else. Judges are often sensitive to public policy around access-to-justice and fairness of the justice system as a whole, but some systems tie their hands in these matters more than others.

Lastly, public funding (and some forms of quasi-public or private speculative funding) of parties who cannot afford litigation with a much larger party is fully in the realm of politics. Changes in government often lead to substantial changes in the resources available to such parties, and there is very little a party can do if it cannot afford to carry its claim or its defence to conclusion in a litigation. Conversely, making it very easy for small parties to bring non-meritorious claims against large parties just because of the apparent size-difference in the parties is also unfair: nobody should ever be able to succeed in pursuing an unjust claim. Preventing governments from tilting the playing field in various directions requires political will which is rarely there, however.

Finally, there's a diversity of opinions on how to deal with the costs of litigation (or even if there is a problem that needs to be dealt with at all) within the legal profession itself. With consensus being fleeting and rare, the lack of political will for governments and legislatures to make changes in how costs are dealt with is somewhat understandable. These differences of opinion are not all motivated by professional self-interest, either. Perverse outcomes (that are bad for the reputation of the system of justice as a whole) are a risk from any sweeping change to the current system. :-(

[1] https://www.justice.gov.uk/courts/procedure-rules/civil/rule...


This is also important in the context of the EU/Brexit negotiations. There's a large faction of UK politics that is upset about submitting to the jurisdiction of the ECJ, but on the other side this gets seen as a signal of the desire to break trust and do things the ECJ would rule against.


>Actually, I wonder if a major part of the decline of trust in the U.S. today is because the price of litigation has spiraled outside of the means of most Americans.

I think the problem is on the other side. Defending yourself is too expensive. Instead of doing quick and efficient business people invoke massive amounts of overhead creating liability reducing paper trail for everything rather than just doing honest business and lawyering up if they have the bad luck of being sued.

It's like paying $100/mo for insurance because you can't afford to pay a $400 6mo premium.

It doesn't help that most people don't have and understand about how tort law works.


This is pretty wrong. Hierarchical relationships exist all over the place, and it's still generally fine to trust your boss, or your teacher, or whatever. And in fact you can't run any kind of large organization without hierarchies, and if everyone defaults to not trusting anyone with power over them these organizations would all fall apart.


I don't think most people trust their bosses, or HR, and for a good reason. For example: https://news.ycombinator.com/item?id=15018007

And I think yes, larger hierarchical (actually, authoritarian would be a better word - you can have hierarchy with democracy too) organizations are inefficient, so in a sense, they have already fallen apart.

I think trust implies choice in some sense, and if you don't have a choice, then trust is meaningless.


People seem to sometimes think of trust as a binary value, but I think if we look at trust in practice on computer systems we see how it's much more nuanced than that.

Trust is an interface. You need to give others just enough trust to do what they need to do, and that differs depending on their role in your life.


Just because the predominate form of organization is currently an authoritative heirarchy doesn't mean that is the only option. See Democracy at Work. I for one don't submit myself to authoritarianism and as little heirarchy as possible in my career.


Be careful of that. Lots of people have come to grief because they assumed because their boss, teacher or whoever seemed friendly, it was okay to confide in these people about some personal vulnerability such as a health problem, the way they would a friend. And just because society needs a nonzero degree of hierarchy, does not mean it must or should have anywhere near as much as it currently does.


I think it's about being able to walk away, not power.

Simple example: Starbucks is a corporate giant with vastly more power than me. Still, I can easily get my coffee from many other places.


> I think it's about being able to walk away, not power.

It's the same thing, though. In any power unequal relationship the power to walk away breaks it.

Examples: In a marriage, if you can leave, then your partner cannot abuse you. At work, if you can strike without worry of being fired (and jobless), then you can negotiate better salary and work conditions. In geopolitics, Ukraine cannot cease to be Russia's neighbor and reinstate itself on other hemisphere of the planet. It a dictatorship, you cannot make the leaders to walk away from you, like you can in democracy.


Seems like we agree on the facts, and disagree about if "power" is the right word.


I used to work for Apple a long time ago and this was one of their 10 "core" values (and I'm sure they intended the pun). For whatever reason, this one value really stuck with me because I'd find that people tend to be more honest with you when you assume positive intent. Instead of trying to defend their actions, they explain what their thought process was in making decisions and taking action and are much more likely to admit when something fails. Between "assume positive intent" and "failure is an option and, sometimes, it's the best option", I find that interactions with people are much smoother and far more genuine.


This doesn't work for me. I assume so much positive intent that I give people a lot of chances without criticizing them. Even when their story makes no sense, I try to give everyone full benefit of doubt. As a result, they seem to think that I am stupid and try to repeat the same behavior.


Those are two different things, I think. Assuming positive intent doesn't say anything about giving people chances. It just means that, when someone does something, you should assume that they didn't do it to be malicious. Multiple instances of something means that they're either incapable of doing what's been asked or they have now shown/proven that their intent wasn't actually positive and your assumption was wrong.


I literally saw this headline, and immediately went to the comments to search "Apple" to see if anyone had already written this (another Apple alumni here).

There's also an aphorism with a similar but different sentiment, Hanlon's Razor [1], that states: "Never attribute to malice that which is adequately explained by stupidity."

[1] https://en.wikipedia.org/wiki/Hanlon%27s_razor


I think assuming positive intent actually often aligns with https://en.wikipedia.org/wiki/Occam%27s_razor . Unless you are seeing reasons to assume any negative intent, it often makes sense to assume positive intent.


I didn't write this article, though I posted it. I've found this to be true in my life. If you assume positive intent, that engenders trust, which can actually increase goodwill. The same is true in reverse.

That doesn't mean you should be a sucker. But assuming the best of people has made me happier.


I agree in general. People act up at the level you expect them to, it is human nature.

But I also think that evidence comes far before being screwed. Watch how people talk about others. Watch how people treat others. Especially people they don't like. And always remember that you might be one of those people they don't like someday.

Focus on building relationships with people who treat people they disagree with or dislike well. Trust those people. Plus, encourage that culture -- no matter how much power you have you should always treat others with respect and fairness.


I'm glad to know other people have this experience and find ways to share it.


> Marth Stout claims 4% of the population falls into this category. In short, if you start from a place of positive intent, you are going to get screwed over by at least one out of 1 out of every 25 people you encounter because they literally have no moral compass.

Just a pet peeve of mine: If p% of the population has a certain trait it means that of 100 people, p will have the certain trait on average -- and not at least p of 100 will have the trait.


True. I'd also point out that bad but rare events can potentially have more impact than frequent good events. It all depends on how the expected value turns out, as many gamblers and traders have discovered the hard way.


I believe this is good advice, but I think it's just a subset of positive thinking in general.

(Most) humans have very well developed subconscious human attitude detectors. We respond to body language, tone, vocal tension, etc., and we often do it without realizing it.

As an actor in a situation, if you can keep your thoughts focused on the positive outcomes - especially when a lot is unknown and your chance of being wrong is not obviously very low - you will express positive signals. Other people are likely to pick up on these and (perhaps subconsciously) respond in kind.

The reverse is very true, and much easier to observe. If you pay attention to other people having conversations, it's fascinating to see how one's initial approach is often met in kind by the other - whether positive/positive, or negative/negative.

Salespeople learn this, perhaps mechanically, very early. Approach a potential client with an expectation of success and an expectation that the client will want and need what is offered. The prospect picks up on this and responds with more interest, or at least a bit more consideration. Likewise, approaching expecting rejection is more likely to get rejection.

I don't have sources for these "facts", but they are echoed in many books I've read and in my human interactions I've witnessed or been involved in.

Lastly, it's just nicer for yourself if you believe people are good and of good intent. You carry less stress, you smile more (and frown less), and you are less likely to die of heart disease. This is clinically proven :).


I love the concept of positive intent since the day I found out about positive intent through another blog post: The power of positive intent[0]. I was always defensive whenever I have to confront with a situation or a person. After I read the article about positive intent, I have been trying not to be defensive and instead, listen to another person first. I think this article "Assume positive intent" really complements the one that I read before.

[0] : https://m.signalvnoise.com/the-power-of-positive-intent-5d5d...


> It’s 100% reasonable to have a high degree of skepticism within a low-trust environment.

And this IMO kills it in an increasing amount of work places. I might be seeing it more pessimistic as it is, but the increased people turnover, team fluctuation, miscommunication and related factors produce a massive drop in trust between co-workers.

> Or we can start from a place of positive intent right from the beginning and keep it there until they violate that trust.

Additionally, I would search for a way to tell people I work with about this principle. Some people begin to distrust you if they think you are kinda odd


Well-said. I worked at a place that had layoffs awhile back and it was a real bummer, in part because of the layoffs, but also because management lied about the reasons for them and the health of the company at the subsequent all-hands. Once management switches to "lie openly and often" mode all the interpersonal behavior incentives change. Like you said, it becomes reasonable to be skeptical and distrustful...


> 1) It’s 100% reasonable to have a high degree of skepticism within a low-trust environment.

The problem is that people aren't able to accurately judge when they are inside a low-trust environment when using one of their many digital devices.

People are watching their neighbors return from the mysterious Facebook fog spouting fake-news bullshit like it's gospel. Then they themselves blithely enter the fog with the reasoning that they would never get taken in by something so crude and obvious.

If you've ever watched a shell game in a big city that fallacy will be familiar. You watch a frustrated victim keep losing and notice a subtle thing the scammer is doing to cheat. You then place your own bet and choose the shell where the cheating scammer inconspicuously tried to move the ball.

Then you lose, too, because you didn't consider the possibility that the scammer had yet another trick prepared just for people like you who think you're smart enough to figure out the game.


What do you mean by "fake news"?

Is it yellow journalism? Right propaganda, left propaganda? Articles from the perspective of the skeptics, from the alt-right, Neo-Nazi's, communists, cultural Marxists or from radical feminists? Something else?


This is terrible math: [...] con artists and sociopaths out there [...] 4% of the population [...] you are going to get screwed over by at least one out of 1 out of every 25 people [...]

On the positive side, we try to put con artists in jail, so that moves the odds a bit in your favor. But con artists and sociopaths continually have to find new people to exploit as they are discovered by and/or burn out the old ones. And the skilled ones can gain disproportionate control of resources, further increasing the odds you'll be dealing with them. As an example, 99.7% of my unfiltered mail is spam, not 4%.

I think you should almost always act as if you're assuming positive intent, because that's how you build good relationships. But in practice, you should always be keeping an eye out for signs you're dealing with one of the many exploiters of the world.


> we try to put con artists in jail

or elect them into high offices


I guess the upside is that we can keep an eye on them.


Ironically enough, the very next headline on HN at the time of writing is 'the incredible shrinking airline seat.'

You should certainly assume positive intent (or one of the many variations discussed here) when you first encounter a problem or conflict of views. On the other hand, you shouldn't cling to this idea in the face of accumulating evidence to the contrary. Some people are in fact assholes and exploit the kindness, patience, and general good nature of others in pursuit of their own selfish ends.

The sad fact is that dealing with assholes is tiring at best, stressful in general, and can be scary at worst. If the cost of accommodation seems less than the cost of confrontation, making excuses and allowances for bad behavior can become a habit. Unfortunately, so too will the bad behavior under those circumstances.


There is something to be said when you are the recipient of the assumption. For me, I would make every attempt to fulfill that assumption. This is my experience when extending the assumption of positive intent. It is implicit trust. When you start from a good point, it is more likely to continue. Especially in business dealings. There is a time and place to keep a distance, but on new encounters and dealings, it is best to move forward.


call me an asshole but instead of trusting or not trusting people, I trust that someone will always look out for themselves first and others second. I'm never surprised when they act that way and pleasantly surprised when they don't.


The problem with that is that it becomes a self-fulfilling prophecy. If you assume bad faith, you're more likely to act in such a way yourself, setting a low standard for the foreseeable future.

A lack of trust will also restrict your options. You'll spend more time making sure all contracts are watertight, or have to pass on opportunities if the legal system (which is the fallback option after 'trust') isn't capable to compensate.

It's also somewhat wrong on the facts: The idea of completely rational and selfish behaviour is called 'homo economicus', and it's among the worst of all the bad assumptions economics has made. Because as it turns out, it's almost impossible to find people acting in such a way outside of mental institutions.

The only rational and selfish test subjects are usually monkeys. Humans are willing to forgo all sorts of rewards for either altruistic or moral reasons.


The issue is not really so much if it makes you an asshole or not. The issue is whether it makes you effective or not. Will you (and other people) be better off if you change your behaviour? Personally, I don't know the answer to that question, but I think it is a very interesting one.


Everyone assumes that everyone looks out for themselves first. That is not a controversial or novel idea.

The question is whether, in a team situation, one assumes that others are looking out for the shared interests as a group. If no shared interests exist, then the team has a much bigger problem.


First, thank you. Whether you agreed or disagreed with what I had to say, I'm honored that you would take the time to read what I've written and to speak up to add your own perspective. I've never had an article get 13,000 page views and an experience like this really helps provide a lot of motivation to continue writing.

Second, I completely understand that I could have made this article even better. I could have done more research into the existing literature, which would have resulted in the Wikipedia entries and company culture documents that were specified. When I take another pass at this article, I'll add many of these at the end for further reading. However, I will say that I in no way am claiming this to be a unique thought. In fact, I'm very clear that this was an experience I had during my career that turned out to be quite transformative. Had it been said in another way, I may not have heard it. It was this exact phrasing that both struck a chord with me and it has resonated with others (as it has done here as well). To that end, even if I could have refined this further and made it 10-20% better, it achieved its end goal.

I honestly wish I had time to review and respond to each and every one of your comments, and I will probably get to it here and there as time permits. I've made some life choices where I prioritize time with my wife and daughter after work hours, which is why I need to sneak in my writing when and where I can! That said, this is encouraging, and I thank you all once again.


One of Wikipedia's main behavior guidelines is a fairly similar idea, "Assume good faith": https://en.wikipedia.org/wiki/Wikipedia:Assume_good_faith


Yeah–I was trying to find out what the difference to "good faith" is, and from a bit of googling, it appears to be an attempt to sell the old concept with a new name?

I somehow prefer "good faith", even though it's meaning is harder to parse when you first hear it. The 'faith' does a better job of hinting at the emotional component of the concept.


Oh come on — you are violating AGF in guessing about the "attempt to sell…". It's far more likely that this "assume positive intent" idea and post is just by people who were unaware of AGF as an existing principle. Follow AGF and just help point them and everyone here to the worked out Wikipedia concept. No need to assume bad faith of the people writing "assume positive intent"!


This is 100% off-topic and just about WP, but something just occured to me: I wish Wikipedia had a way to make new articles as stubs and mark them as such, so they don't show up on the main page, unless people are logged in and want that. Kinda like "showdead" on HN. Even better, make links to stub articles in other articles not be links either. I can't count the times I wanted to add something, but either didn't know enough about the subject or was too lazy to submit enough on the first try. I don't mean to sound entitled, but lowering the bar for "casuals" would be great, if it could be done in a way to not diminish quality (i.e. it would still have to be stubs about something that belongs in WP according to the rules, but not necessarily stubs of good quality). It's a bit disheartening to be shot down by deletionists who aren't even willing to give it some time.. if it can dishearten stubborn me I'm sure it disheartened a lot of people over the years, and who knows what articles could have been grown from tiny kernels by now.


Hah — when I read "This is 100% off-topic" I assumed you were making a completely absurd 100% wrong point about "this" being the AGF link from the parent comment. But you meant "This [following stuff I'm writing about]…"

It is SO easy to misread things, which is one of many reasons that AGF is so important a principle.


"Trust by verify" is always useful. In addition to the caveats you usually have a buffer (of time). A well placed email or phone call can go a long way to clear up (or confirm) any assumptions.

Fighting or fleeing without verification is typically going to end badly.


I think that 'negative intent' is equivalent to 'being stupid'.

The other thing is that 'positive intent' does not necessarily appear as 'positive intent' to everyone. Today, out walking my sister's dog I/the dog scared a Chinese tourist. There was not any malice in the dog's walnut-sized brain, just a perception problem from one of the many (maybe 25) people I met en-route.

There are also those with some narcissism due to childhood trauma and negative feedback loops. These people don't have friends and don't think to do people good turns. Unless they meet a fellow narcissist that sees the world like they do, they can only expect negative intent from people. Of course the world of sane people only see negative intent in those far up the narcissism scale. It may take them a few seconds, weeks or years to realise but eventually the 'utterly selfish' conclusion is realised.

I spent a lot of time in retail, on helpdesks and other customer facing jobs where it is commonly expected that there will be customers wanting to vent, rant and insult you. This never happened to me, it really has been happy customers all the way, admittedly with a few people let-down with promises not kept.

What surprises me is how rare my 'never treated like scum' experience has been. People with more charm and better people skills than myself have had horrible times with customers and kind of expect it. So if your mindset is to consider customers as 'idiots' then 'idiots' you will get. If you start from the 'assume positive intent' and are happy to live in a world of different abilities where you might have to teach someone 'how to use a mouse' or 'type their own name' then all is good and there is no need to be irate with anyone or for anyone to be irate with you.


I used to think this but then I was required to work with two people who have what I understand to be Borderline Personality Disorder. No matter what you think you can do to get along with people with this problem you can't--they will tear you down. The only way to win is to get completely away from them and not have to interact with them.


Certainly some people with BPD are as you describe, but personality disorders exist on a continuum and painting everyone with the same brush is counterproductive. Some people with BPD are lovely people who can be difficult at times.


Remember, it's just as important to follow such a rule yourself, as it is to admonish others to follow. Let's not fail to hold ourselves to the same standard of behavior we expect from others.


Absolutely. In fact, I'd say it is more important to follow it yourself, because:

   1. you can exert more control over yourself
   2. it will have more direct positive impact


I like this rule because not only does it seem to make me and the people I interact with happier, when it doesn't work (rare), I get to enjoy my warm and fuzzy moral high ground feelings. Like my dad taught me, you're never wrong to do the right thing.


I think you are better off if you assume self interest and you make sure that the self interest of others serves your self interest.


Assuming good faith is slightly different than not assuming bad faith, but it's related. Self-interest and good faith are compatible in many cases. Assuming good faith is a principle worth following as a guideline, and when you recognize that it is in conflict with someone's self-interest then you can be reasonable to at least recognize the likelihood that self-interest wins out.


> Chris could have crushed me, and yet he didn’t. In fact, he did the exact opposite and taught me an incredibly valuable lesson. Amidst the bickering on one phone call, he asked his colleagues to stop this behavior and to assume positive intent instead.

I think we've all had this feeling - this reassurance - from time to time. It feels great and it's important to internalize. However, with today's economy existing against the backdrop of the seemingly unstable sociopolitical dynamics of the USA, it's a little hard to know when it's truly OK to operate with this as a key assumption.

> 1) It’s 100% reasonable to have a high degree of skepticism within a low-trust environment. For example, I would never assume positive intent and allow my daughter to be alone with a registered sex offender just because the person claimed they had changed. I would also never trust an alcoholic with a house full of liquor.

This is arguably an exception large enough to swallow the rule in some very important and timely circumstances.

The state is as likely to be violent as a registered sex offender, and as likely to pillage as a drunk at a liquor cabinet. In our current situation in the USA, with the state taking on a character of such ubiquity, aren't we always in a "low-trust environment?"

On the other hand, I do always assume (and in fact, almost always find) good intentions from the humans around me. But we are faced with a real need to come together and address what increasingly seems like bad intent on the part of the state, even though its actors are all humans.

How then do we move forward?


This article argues that you should assume positive intent in lieu of negative intent. I'd suggest that you remain neutral and avoid making assumptions of any kind.


Why?


Because:

1) your assumption biases further thinking and perception. 2) you will be less careful (optimism bias) 3) if you're wrong there might be a big price to pay. (Including the ultimate such as life.)


Sufficiently advanced incompetence is indistinguishable from malice.


I am still waiting for "Presume good faith" to be added into HN guidelines, and I think it would help the HN community improve in terms of quality of discussions:

https://news.ycombinator.com/item?id=12173804


The (iterated) prisoner's dilemma is a massively simplified but nevertheless (therefore?) interesting topic. It has captured our human imagination for decades, for better or for worse:

https://www.wired.com/2004/10/new-tack-wins-prisoners-dilemm...

I say "for worse" because as with all oversimplified models (aka all models), there's always the risk that someone will take the results out of context and claim "But it's a fact! Science says so!". ...As they did with the original "tit-for-tat" IPD strategy, and as happens routinely with various "evolutionary justification for X" arguments.


I actually think there is an extra level to this which is try to understand what people mean rather than way they say.

Way to often discussion ends up in "well you said" instead of actually trying to understand what people are saying.

But Hanlon's Razor already take care of the articles subject.


Assume positive intent, but set boundaries. The first part sounds like wishful thinking but it merely reflects the fact that there are no bad intentions, only mistaken ideas about what is good. The second part reflects the fact that some people know more than others.


There are objectively bad intentions. For example extreme egoism (intent to gain at any cost) or pathologies. (Criminal notwithstanding.)


No I think that people with extreme egoism genuinely think that loving only themselves is the right thing to do.


> Well, there are two different ways. We can start from a place of skepticism until someone has proven themselves worthy to be trusted. Or we can start from a place of positive intent right from the beginning and keep it there until they violate that trust.

I've worked for years in both France and the US, and I feel like it's been a useful simplification that the French start not trusting you until they do, whereas Americans will start trusting you until they don't.

I've heard it described that the French have the latin attitude (shared with Spain and Italy, not with Germany or England), and it's a fun enough explanation that I'm holding on to it :)


This only works if the confrontee assumes the same. Which is often not the case. People usually think others can read their mind, and don't communicate their issues sufficiently well. The natural response to that lack of information is to make things up to have a coherent mental model. Over time this model tends to diverge from the ground truth, sometimes exponentially. As a result you get conflict, assumed malice, and other kinds of misunderstanding. The frustrating thing about all this is without grounds to unconditionally assume benevolence of the other party (i.e. without being close friends), this is largely unavoidable.


I first came across this phrase a decade ago in Forbes as the best piece of advice Pepsi's CEO ever got. It's one thing that has always stuck with me. We often let our own insecurities misrepresent other people's intentions.


Looks to me, the winning strategy. Various strategies around trust have been very nicely gamified here: http://ncase.me/trust/


He says 1/25 people in the general population are sociopaths.

It would be interesting to know whether that ratio changes as power increases. Maybe execs are closer to 1/10th because sociopaths would both be attracted to control and their lack of scruples is a great advantage.

Of course, if a lot of people in power are sociopaths, than that's a huge drag on any social system assuming positive intent.


Sociopaths are violent, the coloquial use is actually reversed from the clinical use. Psychopaths are what most people who use "sociopath" mean. Sociopath != social psychopath. Sociopaths are 100% always violent and zero empathy. Terrifying people. Stab you for your cell phone scary, not politically outmanuever you or work in sales as a profession scary. Psychopathy is the spectrum in question, and it does increase as you get higher power jobs. The recipe for psychopathy would be high cognitive empathy, low emotional empathy, high need to control environment. A low IQ person with those characteristics might be violent or not, and probably adapts to those traits pretty poorly socially speaking. A high IQ person with those characteristics would figure out how to use them to their advantage. It isn't that you have to be a genius psychopath to be successful, it is just that being a genius and removing the constraint of caring about harm to others certainly helps.


Interesting, in the colloquial use, I kind of thought it was exactly the reverse, maybe it's the colloqiual 'psycho' in 'psychopath.' I think most people use this colloquially in the reverse, thinking the 'psychopaths' are the dangerous ones.

Hmm, this article suggests that neither one is a technical DSM term _and_ that it's the reverse of what you say, psychopath is "worse".

http://www.webmd.com/mental-health/features/sociopath-psycho...


Interesting. Doesn't seem that way from wikipedia: https://en.wikipedia.org/wiki/Psychopathy#Sociopathy


You might enjoy reading the Gervais Principle [0], which discusses the social strata in many organizations. One might not agree with it, but it's still interesting reading.

0: https://www.ribbonfarm.com/the-gervais-principle/


It's worth warning though that the labels (losers, clueless, psychopaths) of Gervais principle do not correspond to normal usage of those labels. They are just a metaphor of class within the organization.


Not to mention that the analysis ignores selection and self-selection biases in all organizations. Nor the fact that you might be driven but not have the means. Also mixing up short term with long term in "economic liberty".


Yes, the ratio changes. In this book are some statistics:

https://www.gitbook.com/book/hintjens/psychopathcode/details


By nature I assume negative intent and over the course of my life it's probably cost me and possibly others, a lot. But then a small percentage of people think the best of everyone and end up paying for it with their lives.


This is also known as the principle of charity:

https://en.wikipedia.org/wiki/Principle_of_charity


OK, so I am guessing they stopped arguing during that meeting, but what was the resolution for the project? Did it get completed?


The road to hell is paved with good intentions.


...said the Wolf to the Three Little Pigs.


…except in nearly every case there was no Wolf, it was just the Boy Who Cried Wolf.


...in which case it would be folly to "assume positive intent" on the part of the Boy.

Stipulated: undue suspicion is as much a failure mode as insufficient suspicion. Both hazards might be implicated in any particular situation. That's why it's a mistake to make hard and fast rules. A sense of proportion is valuable.


Assume Good Faith (I'm not going to go with the "positive intent" version, I'm sticking to the known, better formulated Wikipedia principle) doesn't mean you should blindly truly assume it literally, it means that when you interact with people, you should consider how they could be saying what they are saying with good faith and engage in the discourse with that assumption.

AGF is a rule in the sense that if you don't know someone's bad faith, you must act as thought they have good faith as a rule for maintaining healthy discourse. It does not mean you assume good faith to the point of allowing yourself to be vulnerable or agree to anything based on that good-faith assumption.

In short: Assume Good Faith in how you communicate, but verify good faith with certainty if your decisions rely on it.


Nice retort. In the modern fear fueled environment the cries of "but what about the bad guys?!?" have gotten very old to me. Letting a minority ruin your ability to trust people is very unhealthy.


... or do what I say, not what I do.


The comments disturb me:

> Surprising you didn't talk about trusting your gut.

>> I will admit that I struggle with finding a balance between brevity and comprehensiveness. You're suggestion is absolutely right in that I could have put some additional guidance on using common sense and/or trusting one's guts/instincts when something doesn't feel right and it makes more sense to be more cautious.

This seems to be a disturbing trend recently. Inflating people's sense that their biases are not only valuable inputs but actually accurate. This flies in the face of all quantitative studies that show exactly the opposite.

I think the most egregious thing I've seen recently along these lines was a Sam Harris podcast about "the gift of fear"

https://www.youtube.com/watch?v=Uh9OpbJXOHA


Somme thoughts are stereotypes and not biases. As opposed to biases stereotypes are widespread averages of observations and likely to be on average accurate... Which is course does not mean they will be accurate for any specific case. There is such a thing as prior knowledge.




Applications are open for YC Winter 2020

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: