See Allen's comment in the linked post, for example. It's direct ("I'm confused"), but polite. It's asking a question of the submitter in a respectful way that's likely to engender a productive conversation as opposed to putting people on the defensive. Allen's leaving the possibility open that his assumptions are wrong (and often our assumptions are).
It quite literally requires less effort - Allen didn't have to expend the extra effort to type out "this is stupid".
I guess I don't see what's so difficult about that particular type of kindness.
It takes a level of emotional intelligence to communicate in a way which gets to the core of the issue while not making the other person feel bad about it. It can also be a respect thing (when you think someone is wrong first tell them how they're right). This leads to people feeling included and creates higher bandwidth conversations about whatever the real issue is, it also makes people feel more comfortable in general and happier as a part of the group.
I think when people behave like Linus (or executives/founders sometimes do) it's a tantrum/status play. You can only behave like that if you're in some position of power, otherwise people will just dismiss you. This is too bad since people in those positions have a disproportionate ability to actively choose to act against their status (things like Zuck sitting with everyone or in a glass office everyone can see into) which then flows downstream culturally.
Cultures that act 'unkind' or without empathy are a shame because they drive away people who can't deal with the abrasiveness (which often makes things worse), they make people afraid to ask questions and ultimately hurt the group.
You can be direct and empathetic at the same time. If you're in a position of power you have a responsibility to not behave like an ass (or at least you should).
I don't know. Sometimes the core of an issue is that someone messed up, bullied someone, didn't listen, or something along those lines. In some cases people should feel bad. We don't have to go out of our way to make them feel bad. They did it to themselves. And, to be clear, they shouldn't feel condemned. They should feel like they have a productive path forward. But glossing over important issues to avoid uncomfortable feelings isn't really kind in the long run.
Empathy is admirable and humanizing, and maybe we are short on it, but I'm not sold that it's always kinder.
That's basically the idea of 'radical candor' (https://www.radicalcandor.com/wp-content/uploads/2016/03/rad...) and I think the main thing you're getting at.
For example, when a relationship needs to end (a romantic break up, a firing, etc.), there's a limited amount one can do to make someone feel good about themselves. To some degree our challenge in that sort of conversation is to limit our empathy for the purposes of doing the right thing.
In other words, when someone phrases something reasonably kindly and it's not taken well, we can't blame the speaker for the listener's hurt feelings. At some point, even if they're hurt, it's the listener's fault. That judgement can seem cold and maybe counter-intuitive. However, on the contrary, that's just an example of how empathy gets in the way of what is right.
Though I think those cases should be the exception to the norm.
As far as empathy getting in the way - I think in the firing example while managers often feel an urge to be empathetic in a misdirected way (i.e. "this decision is hard for me too") true empathy would be recognizing that if you were the one being fired there is no good way to state it and however difficult it is for you to fire them - it doesn't compare.
I guess to reframe it, it's more about thinking about how the other person will be affected and doing the best you can given the circumstance.
No, the empathetic thing in those situations is to end the relationship quickly precisely because you understand dragging it out will just damage the other party more in the long run.
No - empathy just means feeling the same emotion as people around you, which is not always helpful. What you want is compassion.
I often see "empathy" tossed around as a cure-all for social problem. When someone gets angry at work and it makes you angry - that's empathy. Ever seen footage of a frenzied mob saluting a fascist dictator? That's empathy as well.
That's a particularly narrow definition of empathy, most definitions include something like the following:
"the ability to understand and share the feelings of another." 
The capacity for this is what helps - it makes it easier to understand how the other person is feeling.
As opposed to compassion:
"sympathetic pity and concern for the sufferings or misfortunes of others." 
Empathy is a more general understanding of emotional state - I don't think it's being incorrectly used.
The OP suggested merely leaving open the possibility that your assumptions are wrong. That's really the opposite of understanding, it's expecting that you're not always going to understand, and making space for the other person to explain themselves and make things right.
Hmm, seems more like you just have to assume that people aren't stupid, and there's a reason they believe the things they do. Until you understand that reason, you just have to keep saying you're confused and ask questions to further clarify.
Once the reason(s) is highlighted, sometimes it's well justified, sometimes it isn't, but at least the core issue can now be addressed.
There are plenty of opportunities to do this when giving criticism/feedback. Not only does it make it easier for the other person to hear, but it also helps you save face if your feedback ends up being wrong.
Rather than "blame", I suppose we're to say Allen accepts responsibility for his confusion and makes an effort to enrol others in his attempts to seek clarity. That way the opportunity arises for any involved to say "ah, I see, thank you".
If we hear what people are saying with a sense of goodwill, and read their words charitably, we give less cause for animosity to arise.
FWIW I don't see this as placing blame on himself, at all: I see it as Allen just stating a fact. Allen is confused and seeking clarification. Confusion can happen for all sorts of reasons with varying parties or circumstances being the cause.
Allen might similarly say "I'm confused" upon reading the time cube website, but Allen's not to blame for that.
I agree with you though that largely politeness comes from being internally focused ("I don't understand, I don't agree, I didn't know") as opposed to externally focused ("You did something confusing, You did something wrong, You didn't tell me"). One instance where it pays to be selfish, I suppose!
It's not about "difficulty" ... as in one takes 5 calories to type out "that's stupid" but the polite sentence takes the same effort or less.
It's about effectiveness and possibly the false cause & effect we attach to one style of communication vs another. E.g. Linux may be successful because Linus is direct and blunt. (It keeps idiots wasting everyone's time away.) Or, Linux is not reaching its full potential because Linus is crude. (It drives talented contributors who are repulsed by crude language away.) We don't know which scenario is superior because we can't replay history in 2 separate universes.
To your particular example, I personally don't like the "I'm confused by this." That is what I call unnecessary "prelude". It's redundant prelude because Allen's next paragraph clearly asks 2 questions so of course he's confused.
Also, Allen writes:
"Can you take a step back and explain your goal here?"
Allen can delete "I'm confused" and "take step back" and get directly to the point:
"Can you explain your goal here?"
That holds especially true when your job is to provide that broader picture for others, and the person you're talking to might not have your perspective.
Having had to give the equivalent of the "No, no, I meant" follow-up response there more than once, I've learned to provide the additional signals to up-level.
Then it depends on how we perceive word choices. To me, "take a step back" has become so overused that it's a "dead metaphor". It has become the opposite of "intentional phrasing" and its most common use is prelude and lubrication.
I'm not saying you're wrong in interpreting "take a step back" more literally. (Literal interpretation is the premise of what your comment dissecting it is based on). Most of the time, you can ignore it based on the way most people insert "take a step back" into their business-speak.
Also, I doubt that surveying 10 random business people what "take a step back" means would result in 6+ out of 10 defining it as "higher-level concerns". It seems like "take a step back" acts more like a discourse marker.
The higher-level point: you're trying to produce a certain reaction in the reader/listener, and every word and phrase you use will contribute to that reaction. Even if you don't care about how they feel for its own sake, there's value in producing the reaction you want and avoiding the reaction you don't.
So, in my case, I've had direct experience with specific choices of phrasing producing undesirable results (and in a way being insufficiently direct), so I've adopted other phrasings that produce the results I do want. The specific things that work for you may vary.
> Also, I doubt that surveying 10 random business people what "take a step back" means would result in 6+ out of 10 defining it as "higher-level concerns". It seems like "take a step back" acts more like a discourse marker.
In the specific context of the whole phrase ("Can you take a step back and explain your goal here?"), I'd consider it semantically meaningful. I wouldn't be surprised if it gets used in semantically null ways, as well.
But even if it were simply a "discourse marker", that shouldn't be automatically discounted and deleted either. To make an analogy to another form of behavior: I've seen a lot of hackers follow a certain counterproductive chain of reasoning regarding fashion. "This makes no sense and shouldn't be a thing that matters. I'm going to treat it as a thing that doesn't matter." (It's a form of the is-ought fallacy.) As opposed to the rather more effective reasoning: "This is a real social signal that turns out to matter, even though I don't think it should. It costs me little to play the game and obtain the results I want."
I'll freely admit to using empathy both because I care about other people and because I've in practice found it incredibly effective.
What is a discourse marker? Not that I have a problem with the phrase "take a step back", I agree it depends on the environment, the context, on the person who is saying it, how they're saying it, and who they're saying it to. But if and when it's just filler, what stops it from turning into "can you go right ahead and take a step back and explain your ultimate goal target here?" If it's a "real social signal" (as opposed to a mere social signal?), what does it signal? Do I want to be signaling that?
There's nothing inherently bad about discourse markers (or dead metaphors) for that matter. They are quite critical to a great deal of communication. I think what the parent was trying to suggest that this meant its use was somehow implicitly deceptive or something, but I don't know why.
Know your audience, and make sure you're sending the signal you think you're sending. If you don't already have all the reputation you need, fashion matters a lot more in establishing a baseline.
So starting off with "Can you explain your goal here?" is concise and isn't horrible. But without any other signal or adornment, it risks being read as the above kind of question. Leading off with "I'm confused" is for indicating that the subsequent questions are the other kind - "please help me with my problem" type deal. It's not strictly necessary to the meaning, but likely has a positive effect in how the questions are interpreted and answered. It's like declaring a variable before you use it. The compiler doesn't need to figure out what type it is (and possibly get it wrong).
"Take a step back" is also used productively here, indicating "whoa whoa there sparky, before we talk about what you're doing, we need to answer why you're even doing it." In other words "I want to talk about something further upstream" or in yet other words, "Let's stop and look at the big picture." Actually it's pretty concise when you consider all the meaning that's crammed into it.
I don't think redundancy or prelude are necessarily the mark of politeness. It's usually just a mark of writing style. For example, you said:
> To your particular example, I personally don't like the
But "personally" is redundant - of course it's your personal opinion. You just used it to emphasize that you don't like Allen's writing style, correct?
But to _your perception_, it's politeness. (You previously wrote: "It's direct ("I'm confused"), but polite.")
On the other hand, I categorized it as prelude.
To attempt a meta observation, it seems people require "social lubrication" in the communication. Otherwise, it sounds harsh. The lubrication is a mark of kindness.
I don't require lubrication and harsh language like Linus' writing style doesn't bother me. The difference is that I would probably use "kinder" language in more situations than he would depending on the recipient.
>But "personally" is redundant - of course it's your personal opinion.
Right... to be meta, it's a restatement of my last sentence. "I personally don't like" is _softer_ than "I don't like" which sounds more like I'm a dictator. If you and I were founder & cofounder where we have "spirited" debate, I'd leave out the redundant "personally". If you were my employee that needs softer language, I put in the redundancies.
Yes, it is to some degree a "social lubricant". We're running a massive intricate machine at full-tilt with a million moving parts that sometimes just barely mesh. And while some parts of it are capable of continuing to work without that social lubricant, it certainly makes things run hotter, and decreases the MTBF. All the more reason to leave some safety margin, and help everything run more smoothly.
I disagree with the view that "softer language" is something you use with weaker people or people who you see as "below" you. I'd also disagree with the characterization as "softer". Empathy isn't "soft", it's incredibly powerful.
That's not at all why Allen's comment is polite, and this is very much what I mean by "kindness requires no effort."
It's polite because it isn't actively negative towards the other person. That's all it takes. Redundancy is irrelevant.
Allen could've easily written:
> What you wrote makes no fucking sense and it's dumb as hell.
Which is both redundant and impolite.
> If you were my employee that needs softer language, I put in the redundancies.
Just to be redundant myself, redundancy is only loosely correlated with politeness. There are plenty of examples of terse, polite speech (and the inverse as well). You don't need to do this, and I would argue it's a hallmark of your particular writing style more than anything else.
If you believe redundancy is irrelevant, we have different opinions on how people perceive communication.
You: not writing an insult is all it takes to be "polite".
Me: People will perceive the omission of prelude and mea culpa "I'm confused" as "active negativity towards the other person". It doesn't require the literal words "you're an idiot".
So I disagree that simply not insulting is "all it takes". It often takes more than that to not make people act defensive. Lubrication, social grooming, etc.
I'm starting to gather as such, and so continuing down this road is probably unproductive.
I suppose we'll have to agree to disagree.
The difficulty is in suppressing knee-jerk reactions and hardwired negative emotions, not on the energy level required.
It was about energy in this case because I was responding to BenchRouter's specific context for "difficulty" when he wrote: "expend the extra effort to type out "this is stupid"". (https://news.ycombinator.com/item?id=14782723)
The kindness and softer approaches are forgotten; all of those times when things aren't in need of a strong shove back to positive territory.
HOWEVER, that preface "I'm confused...[followed by pointed criticism disguised as a question]" often comes off as condescending, rude, and disingenuous. It is often infuriating to the target and those who are witnessing it.
There's always a polite way to say things without masking what you really mean. It may be difficult to come up with the words and sometimes it just isn't feasible, but that's OK, if folks are willing to give each other "a pass" every once in a while and tolerate an occasional outburst.
The secret, I've found, is to assume that everyone is doing their best with what they've got. Everyone has a different set of information, background, experience, etc. Maybe you'll have an opportunity to teach them something; maybe they'll have an opportunity to teach you something.
You should only say you're confused if you are legitimately puzzling over what seems to be a contradiction.
On the other hand, if the other party sees a request for clarification and interpretation it as a passive aggressive criticism, then that's pathological on their end.
Which, taken to its logical conclusion, means that some important figures in our industry (e.g. Linus Torvalds) act quite unprofessionally. Sadly, this seems to be true.
No, it doesn't. It means that the people Linus has to say harsh things do are being unprofessional--because, as Linus himself said in the quote given in the article, when he tries to say things kindly--"Please don't do that"--people don't listen. If those people were being professional, they would realize that when Linus says "Please don't do that", they need to listen.
The article even acknowledges this:
"If you really have trust and conviction that your peers are not wasting your time, you pay attention when they say something - a “please don’t do that”, to quote Linus’ example, ends up being one of the more strongly worded pieces of guidance you can get."
So the problem isn't Linus, it's all the people who don't take the time to realize that it's not a waste of time to pay attention to what he says--so he has to shout to make himself heard.
Build a community that doesn't have that problem, and people will respect disagreement and "don't do that" without the accompanying toxicity.
The Linux community is not built around flames and attacks. Linus' flames are what get public attention, but that doesn't mean they are the majority, or even a significant minority, of all his communications. He has explained many times exactly what the conditions are under which he will flame--basically, that a kernel developer does something that he has already explained many, many times should never be done, like breaking userspace. Do something that meets those conditions, you get flamed. Don't do something like that, and you don't.
I'm also aware of projects to do more formal analysis of various FOSS communities, such as FOSS Heartbeat.
But in the meantime, the plural of "anecdote" is absolutely "data" for sufficiently large values of "plural".
I disagree, but I doubt we're going to resolve that here. To me, "the plural of anecdote is data" means "I can't be bothered to figure out whether the information I have is actually representative of the population".
as a (volunteer) coach, one of the things i often have to correct is making lob passes (slow, often misplaced and disrupting of the flow of action). some kids mistakenly believe they are being nice by giving their teammate a soft pass, but i tell them the kind thing to do is making the direct pass. it gets the ball there faster, it's more likely to be in stride, and makes it harder for the defense to recover or steal. that means it's more likely to lead to a basket and less likely to lead to a turnover for the receiving teammate.
(tl;dr: direct is better and more appreciated than roundabout)
It's like asking everybody to not smell or to not look ugly.
Some people are considered rude for a lot of reasons, mainly cultural, but what they have to say has value regardless
Instead of asking people to be kind (which often means comply to my system of values) we should ask people not to be judgemental and be open minded, which is something that can be learned by everybody
You're arguing semantics here; in almost all cases that is what kindness means in these contexts.
I appreciate the fact that in English I can be more direct without being told that I'm rude though; it happens a lot more when I'm having a conversation in Italian.
I think language shapes the way people think, Italian has so many different and polite ways to be assholes that kindness to me it's more in the intentions people have than in the way they express them.
Any attempt to understand, evaluate or otherwise politely disagree is met by a mob of twitter accounts ready to call you a critical asshole and troll at every turn. I'm not sure why the industry has shifted this way. My only guess is it's generational.
According to who? Certainly not the article: It puts forward Allen's comment as an example of being kind.
> Any attempt to understand, evaluate or otherwise politely disagree is met by a mob of twitter accounts ready to call you a critical asshole and troll at every turn
You could easily change your statement to "any attempt to do anything is met by a mob of twitter accounts ready to call you an asshole" and it would be just as true.
I don't think twitter outrage is a good metric for anything, because it's extremely contradictory, arbitrary, and capricious.
Sure, that's nice rhetoric. And yet the "kind" Bezos has presided over some of the worst working conditions in the developed world  while the "blunt" Torvalds has kept together the very scattered Linux team for decades without controlling their income or work conditions. Apparently the more money you have, the more you can get away with a "do as I say, not as I do" standard.
Grandmas tend to be kind, but these people are definitely not anywhere near that realm.
He doesn't take decisions that can screw thousands of people to his profit. Whether he would if he had that power, he didn't strive to get that power, and he doesn't have it, so while we can't rule it out, we also can tell that he isn't kind with any certainty.
And surely he's not "unkind" just because he's famous or has a full mouth. That's not what kindness is about -- I know people 10 times as foul mouthed, that were real sweethearts and could give you a liver if you needed one.
This is work that improves many lives.
If we had to rely on microsoft and had no alternative could we get to sites that host anti-establishment views? We can never know, and need not go there thanks to these men.
This is a bit of an odd non-sequitur. Are you implying that the World Wide Web (1989) would not exist if not for Linux (1991)? Or that it would somehow be censored of all Microsoft criticism?
I think Stallman is also a kind guy and the world is a better place for having him around.
People say he is this and that, but I think some of his eccentric behaviour is just a survival mechanism.
Would he still be actively working on Linux 26 years later if it had never become popular enough for him to significantly monetize?
I can't read his mind of course, but most open-source founders below the "companies pay you big bucks to work/consult on it" level do end up burning out and abandoning/leaving the project or taking a reduced role after a while. And of course even if he wanted to he couldn't have worked on it full-time, for obvious reasons.
That's of course not to say he hasn't done an amazing job on Linux, or that he doesn't deserve particular praise for the first few years when he really was giving away his best work to the world for free.
So? Who said he was ascetic (or should be)? What we said is that he gave the world his work for free.
And while he might be worth "$40m+" the Linux industry he helped create and spearheaded is worth tens of billions (if not hundreds). Heck, Git products alone should be close to a billion.
He could have easily gone for a slice of that pie early on, create branded proprietary lines of his software, etc, and it would have been a different landscape.
Wow. That sent my BS sensors straight up to saturation. Someone has an agenda here, and it's not to inform you.
As if some country has solved this? In every country I know the security industries are filled with extreme right wingers and/or hooligans.
Steve Yegge didn't call him the "Dread Pirate Bezos" for no reason, you know.
If you're rich or their salary depend on you, you are eclectic
This is one of those statements that I think we want to be true but we have no evidence that it's true. Many contradictory examples exist in the real world:
You can yell at your team and insult them and be successful. (Famous examples are Steve Jobs and Bill Gates' "that's the stupidest idea I've ever heard!")
You can be soft-spoken and be successful. (Warren Buffet would be an example. He doesn't yell at the people in his Omaha office or his presidents/CEOs at Berkshire subsidiary companies.)
Likewise, you can be blunt & harsh and fail. You can also be diplomatic & nice and fail.
Same in other endeavors. You can yell at the football team and win the Super Bowl (Mike Ditka - Chicago Bears). Or, you can be soft-spoken and win the championship (Tony Dungy - Indy Colts). Likewise, you can do either style and still be the worst team in the league.
Doesn't seem to be much correlation either way.
My conclusion based on life experiences is that companies can have both the blunt and the diplomatic approaches. The blunt communication works well in upper management. (E.g. one VP tells another VP that "it's a stupid idea.")
Everybody is a Type A personality and has a thick skin.
However, the reality is that many employees (especially lower-level positions) feel demeaned by direct language. (As the endless debates about Linus' style attests.) Therefore, they require indirect language and those VPs have to dynamically adjust the communication to that personality.
Personally, I don't like the style of indirect communication the author uses in examples of Daniel, David, and Allen but I fully understand it's necessary in the real world for certain people.
I don't actually think the first two quoted comments are the best examples, precisely because they show a degree of indirectness. You can be direct, clear, and professional while still having empathy for the person you're speaking to.
I do have enough personal experience with people of both types to have a strong impression that there's a correlation between empathy and success. By "empathy" here, I'm talking about the ability to accurately model other people and predict how they'll react to certain things. That doesn't automatically imply kindness, it's just a common side effect (since with empathy, you'd be knowingly hurtful otherwise). If you can't model other people at all, you're going to have a hard time in a people-oriented, communication-oriented role, which includes almost any leadership role. You might succeed in spite of lacking that skill, but you won't succeed because of it.
The counter example is mobsters, they are highly successful, have built empires that are competing against the major corporations but are not showing any sign of empathy or mercy,l and I would say most of them cannot even be considered humans.
And for sure they do not understand people, they force them to do what they need them to do, as if they were tools.
I very strongly disagree. You can learn empathy, and teach it. There are whole books about learning to model other people.
"How to win friends and influence people" is, in many ways, all about empathy, for instance.
(Let me distinguish for a moment that there are people out there who have psychological conditions that make it incredibly hard for them to model other people. I don't know how to address that, offhand.)
> But you can't teach empathy.
Yes you can. Social skills can be learned.
In the absence of context for the comments in question, I'll have to be somewhat broad in template. For the first, I'd avoid the passive voice like "It would be nice to...", and instead write something like:
"X and Y look good, thanks. For Z, I'd suggest putting this in a data-bind, so that we can more easily switch the view between AB tests in development."
For the second comment:
"I think this will work in many cases, but it might break some limit in the size of $in clauses, which may lead to problem X in the future. I think we can defer that until it becomes a problem, but please keep it in mind if you see a clean way to address it."
For my part, I find that adding rationale also adds more empathy, because it gives the other person the information to reach the same conclusion you did, and it shows trust that you can make non-opaque decisions and still have them hold up. Also helps with mentorship. And taking the time to talk about the things that are good helps when you then need to talk about the things that aren't; I prefer that over the "no news is good news" approach to reviewing.
Saying "It would be nice to put this in a data-bind" (as in one of the examples in original article) is okay if you think that doing so is optional, insignificant, and the recipient would understand it the same way.
Saying "It would be nice to put this in a data-bind" is okay if you think that the recipient should definitely implement that change unless there are strong contrary arguments, and the recipient would understand it the same way.
If you say the same thing expecting that they'll understand one of the above interpretations, but they understand the other one - that was not empathetic, and you caused a miscommunication. Empathy by definition is personal and contextual, not universal.
I would use very different phrasings for something optional for the reader to consider versus something critical that they should either do or provide clear justification for not doing.
> Empathy by definition is personal and contextual, not universal.
Somewhat true, and you can get better results if you know the person you're talking to better (which is one reason why it helps to meet the people you work with in person, at least from time to time). But you can also have a mental model of a collection of potential people, and choose responses most likely to produce positive results. You can still have certain priors for your expectations of the people reading your words, even with minimal information.
This was one of the biggest things I had to learn in my first management/executive position. Just because all of the senior management enjoyed aggressive debating and sussing out of ideas didn't mean we could use the same language and approach when talking to other employees.
I'm definitely ashamed by some of the emotional harm I caused our lower-level employees.
Also it's important to note that you also need to be aware of who is listening. If you call a fellow VP's idea stupid, that can provoke negativity from more indirect listeners (even though he has no problem with it). Even if they aren't personally the target of direct criticism, many people are uncomfortable even hearing it about others.
I'm guessing a big part of this is how secure each person feels. Someone in senior management already has a highly successful career, they could probably get another decent job pretty easily, and their income is high enough to where they probably have a fair amount of money saved up in case of a rainy day.
For lower-level employees, this is frequently not true. Savings are more often minimal, many would not be able to easily get another good job. So naturally they will feel more threatened by blunt or 'harsh' language.
> they could probably get another decent job pretty easily
This is actually pretty false. Management jobs take a lot more time and effort to find than lower level ones. In tech especially, IC job hunts are way easier than any sort of management roles.
Direct/rude: "Wow, X is a stupid way to do Y. X will break case Z, but I guess you're not smart enough to notice things like that. Start over and try not to waste my time again in your next pull request."
Direct/kind: "I see you're doing X, but there might be a better way to accomplish Y. For example, X will break case Z. Could you try a different implementation with that in mind?"
Indirect/rude: "You must be an idiot to try X. Users who Z will hate you if this change makes it through. Out of all the ways to do Y I don't know why you'd choose this one."
Indirect/kind: "I appreciate your effort, but I'm not sure about X. You may not have realized that Z is an important use case and there might be other ways to do Y. It would be nice to think about this some more."
They even highlighted an example where they say that bluntness can be positive in the article about the study:
> When Rozovsky and her Google colleagues encountered the concept of psychological safety in academic papers, it was as if everything suddenly fell into place. One engineer, for instance, had told researchers that his team leader was ‘‘direct and straightforward, which creates a safe space for you to take risks.’’ That team, researchers estimated, was among Google’s accomplished groups.
You absolutely should not praise people for screwing up. That's not in any way kind.
But you don't have to berate them, either.
I disagree. If you treat people like they are weak, then you will breed weakness.
I have seen first hand what this eggshell walking leads to: projects that should have been shot down as stupid ideas instead wasting million of dollars and hundreds of man years.
You can certainly get a point across by being direct, but to make a truly lasting change you need to convince people it's a good idea. I've yet to see this happen without kindness and diplomacy.
So while the IT security officer can certainly get a strict password policy implemented, without also making sure people understand and agree that security is a good idea the end result becomes a lot of written down passwords hiding on postits under keyboards.
I'm sympathetic, but I bear bad news. You can also just run off, silence, or marginalize anyone that gets in the way. I've seen this happen in corporate politics quite a bit.
It's a skill, most in tech don't have it
(I am certainly no worse at technical pursuits for it, either, and it's one of my deepest criticisms of the way that our culture is expressed is that idea that kindness and understanding somehow make us worse at delivering.)
I hadn't heard of "Radical Candor" before its mention elsewhere in this thread, but a quick look at it turned up promising results, especially the fact that it distinguishes between empathy and directness, and discusses the failure modes of having either without the other.
Not when it really is a good idea
And you realize it somewhere else when they can look up numbers and say: hey this is better than what we had, good for us we didn't assume but we measured.
> Not when it really is a good idea
If you do IT at a company that isn't a tech company, you're usually going to be stuck in a situation where you have to convince a non-tech-head to make a technical decision that seems right to you. Since you don't have four years to teach them everything that got you to that point, you're going to need to give them high-level details.
Getting someone to do ANYTHING that they aren't currently doing requires some level of convincing, regardless of whether its "a good idea" since if the person believed it was the best idea they'd already be doing it.
This is so horribly naive that I worry about your ability to survive in life ...
Most people resist change. Ferociously. Even if it's a superb idea. Even if resisting might get them fired.
Don't worry for me, I'm probably way older than you and I survived just by not fighting lost battles
They resist and then they die
I'm from Italy, I know very well what resistance to change is
Everyone around me does just that and they are literally starving with no jobs, while I'm enjoying (almost) valley level salaries, while also enjoying Italian life style.
> Even if it's a superb idea. Even if resisting might get them fired.
So you prefer to be like them and waste time convincing suicidal maniacs that you're right?
What's the point?
I found this on the front page just now
Employees Who Stay in Companies Longer Than Two Years Get Paid 50% Less
Long story short: if you need to convince someone at you job that your idea is good, unless you're trying to convince them that flying machines are the future and it's 1865, you're better off leaving.
Also, the adage that we all know and has been around since the dawn of time is just as true now as it was at the dawn of man.
You get further with a carrot then a stick. With reasonable people of course.
I'll submit that personal remarks like "only a fucking idiot would..." and such are bad not because they hurt feelings but because they are worthless and distracting. They make the conversation about a person instead of what people are supposed to be talking about, if only for a fraction of a second, and can disrupt conversation.
If someone is doing something that harms the objective, you tell them what they're doing, why they need to stop and possibly how they can fix/improve things going forward. That's effective blunt criticism, and there's no need for personal insults anywhere in the chain.
I agree. If conversation involved two people, both speakers and listeners should make efforts to improve how criticism works. Listeners can help by not taking everything personally.
Coming up with a failed design doesn't mean you're a failed human being. It doesn't even mean you're bad at your job. Learn something, improve, and you'll design something even better next time.
To paraphrase Charles Murray: "nice" is a moment-to-moment tactic for avoiding conflict, not a guiding principle for living your life. We should default to being nice amicable people, but being good often requires otherwise.
Unfortunately, niceness has been raised to the highest virtue in recent years. This is a mistake with civilizational consequences.
I don't think the examples given are examples of kindness.
Concretely, they're insufficiently direct.
If you think someone is doing something that isn't well thought out, and you think you understand the problem well enough to say that they haven't thought through it fully (which is a scenario that arises in workplaces), don't say that you're "confused". It's a variant on false shock. Just say " I don't think this change considers the following scenario:". You can soften that with a disclaimer of "perhaps I'm missing something", but saying "I'm confused" when you think the other person is consumed is mildly passive aggressive.
Likewise, if you think someone should do something, don't say "it'd be nice if we could". Make the request directly. You can still add "let me know if there's something I'm not considering that prevents that". It's frustrating otherwise, because it is unclear what is a request or nice-tp-have and what is an instruction that approval is contingent upon. In the long term, lacking that clarity becomes annoying, especially for non-native speakers or people from different cultures who expect different lvels of directness.
There is a position between aggressive "don't do that, it's stupid" and the indirect formulations in this post, and that's where you should aim. Polite and kind, but still clear and direct.
Honestly, if you just state the problems with the approach clearly and avoid words like "stupid" or "dumb", you're 90% of the way there.
As much as I like the ideas this post advocates, I feel like some of this is on a case by case basis.
It should always be a goal to keep criticism professional, not personal.
One other thing that should be kept in mind here of is culture.
I live in japan where you really can't even say "no" let alone "wrong". There's are extremes like: Linus and the other being many asian cultures.
Like any advice like this, try to look at the intent and the points that work for your situation not "Silicon valley startup only".
An environment of trust (and safety) allows open technical discussions and lets you come to decisions in a way that helps everyone learn and evolve without "losing face" and without breeding an undercurrent of anger and resentment. Knowing that each person is willing to listen to the other respectfully and that each person is prepared to say they are wrong, can improve the discussion rather than making it more wishy-washy.
You need to have this if you're going to be working day after day, maybe for years with the same people. Lose trust and the feeling that it is safe to make potentially "stupid" statements, and people will just blindly follow the loudest most belligerent person because it's not worth the emotional cost of trying to engage in "debate".
So maybe "Trust is Underrated" would be a better title for the original article.
Often, we are nice because we are afraid of hurting people's feelings. As a result, though, we sometimes end up stringing people along and the ultimately make them lose more time and energy than if we had breached their comfort zone early, and communicated our expectations when they weren't yet super-invested. And after all is said and done, if we string them along, they end up blaming us more as well.
This was a hard life lesson to learn, but sometimes, to be kind, one must risk not being nice.
My advice would be: before communicating a tough expectation, do your homework (research how it's done) and be diplomatic. Different cultures have different linguistic paradigms that help grease the wheels towards agreement. Use them. And at the end, be firm but offer support for the transition. If they want it, they will take it. In any case it's likely you will be respected and won't burn bridges that way.
I'd certainly agree that you shouldn't "string people along". That's not kindness. And sometimes you have to deliver news that people will find upsetting. That's a more critical time to apply empathy, not less.
There are several ways to do necessary, un-nice things in a very kind manner. It is a skill, and for some people requires coaching on how to express themselves professionally and kindly most of the time.
(Qualifying this, because there are occasions that warrant a good moderate yell or two. E.g., the 3rd (rejected) and random PR submitted by a Jr. developer that changes all of the spaces to tabs in the repository, or all of the modules/functions to classes/methods.)
It's still annoying that becoming a manager is correlated with advancement, but that's life.
Confrontation is critical when you have important information or perspective that can change the current course or result in a different decision. It's your duty to be assertive (when it's important, not for trivial things).
Being direct doesn't mean being a jerk or being rude. There's no excuse for combativeness. The point isn't to pick a fight or attack a person, it's to attack an idea and to do the right thing. Combativeness is a sign of immaturity or inability to deal with high stress.
It can take practice. It's liberating once you get in the habit of it.
This saying: "You are the average of the five people you spend the most time with," sums up the the importance of work culture (same with choosing a partner and friends too).
> Especially when it seems like no manager qualifies as "kind." So if you want to advance, what do you do?
Stand out by being kind, watch as the people you work with are shocked by the contrast, and try not to get whiplash at the speed at which you're propelled upwards by a large crowd of enthusiastic supporters.
Or, if the culture of not caring about other people seems too pervasive, find a company where kindness is valued, rather than derided. People who value kindness will select for it. And the resulting teams work very well together.
In a technical leadership positions, communication is by far the most critical skill. The job isn't solely to "get your point across". Kindness makes a great impression on people, and builds a reputation. A lack of kindness (or a lack of empathy, or toxic behavior, or any number of other descriptions) makes an impression as well, and also builds a reputation.
Edit: Mildly interesting. 50 minutes after this comment, I was searching for how to calculate the intersection points of collinear line segments which brought me to Stackoverflow ; where someone linked to this sketch  about drawing red lines with green ink, all perpendicular; which then brought me to this sketch  involving politely telling someone to fuck off.
I still don't know how to calculate the intersection points of collinear line segments.
A good workplace culture is, essentially, leveling up from this. It's agreeing while diplomatic language is more comfortable and it's how we might communicate outside work, we're agreeing to suspend it to better achieve our shared goals. If someone challenges your idea, you need dispassionately and genuinely consider their objections and either defend your idea or acquiesce to the better idea. Some people just can't do this. Ideas are personal things and arguing about them feels uncomfortable and they don't like to feel uncomfortable. And, maybe getting a little carried away, but I think there is general societal issue where we think if you're uncomfortable something must be wrong. Good decisions are born out of argument not trust. Saying "I'm confused" or "Help me understand" when you already understand and just disagree is level 0 language. It kinda works but it's slow and inefficient and as engineers - this isn't good enough.
Not sure why "being nice" and "getting ignored" seem like they need to be paired. Linus failed at being nice and not being ignored. He wasn't ignored because he was nice.
My question is now:
Why does it have to be an endless discussion on the merits, vs shutting the discussion down?
You can have a firm discussion on the merits without shutting it down.
That explanation of kindness doesn't make sense. Some people try to be nice and, by mistake, end up being rude. And business people make deals quickly all of the time, using jargon and cutting out pleasantries while still being kind.
No, kindness is a skill of words and actions that must be developed over time. It's about navigating complex ideas and decisions effectively.
For instance, "no" is generally rude, not because it's too short, but because it doesn't provide good feedback on a complex idea. What is the proposer trying to accomplish? What existing alternatives exist, or what others might be explored?
If you don't have the time to give good reasons, then point them toward others that you trust to give good advice. E.g: "This proposal is unacceptable. Discuss with group XYZ and explore alternatives." Or even: "This proposal is unacceptable -- the proposed use case is not important enough to justify what you are trying to do."
The flip-side is that high quality maintainable code is the product of top-notch commits, and rejecting commits is sometimes necessary to keep the standard of quality high. A good maintainer shouldn't cave to pressure of accepting a flawed commit just to avoid hurting someone's feelings.
This article in fact had what looks like a prime example of that. The comment mentioning a PR might "break a limit" but "we'll cross that bridge when we get to it" was touted as an example of how to give guidance. I'd argue that code quality slipped right there as a direct result of social pressure to accept a subpar commit.
It's not easy by any measure, but I think it pays to be not only clever and kind, but also consistent and firm when it comes to reviewing people's work.
In the absence of the broader context, I would tend to guess that that comment was trying to avoid overengineering. We've certainly had more than enough discussions on HN about "move fast and break things". This is the kind of review comment that goes along with doing that intentionally rather than accidentally.
That technical decision might be right, or might be wrong, but either way it can be presented with empathy; empathy doesn't need to change the message. You can say "I think this might cause problem X, but that won't be an issue for a while, so let's deal with it later", or you can say "I think this might cause problem X, you need to fix that before this can go in", and either way you can be professional and kind while getting your point across.
Consequently Linus says he decided to go in the direction of communicating in the manner that he is now known for. (Which makes me wonder-- if he had a personal encounter early on with his sarcasm causing the same bad outcome, would he have decided as confidently to go in the other direction?)
Regardless, I think jaromil who maintains Devuan is a great counterexample. He's quite nice and non-sarcastic, approachable to newcomers, and he seems to be able to herd cats just as well.
Better to let Devuan reach the scale of Linux before attempting to use Denis “Jaromil” Roio niceness as a counterexample to Linus Torvalds.
Even nice people have breaking points. They are perfectly capable of snapping back at you, if you push them hard enough.
In my last job I had lots of hour-long arguments with coworkers on different topics, many of which I ended up conceding the point. I'm incredibly appreciative of them having taken the effort to help me understand the their views, and convince me otherwise.
I think there's a lot of stigma on disagreeing with people. But I don't see why that should be the case. If you have an argument with someone and you both end up leaving with a better understanding of the problem, why is that a bad a thing? I've had plenty of discussions where I fundamentally disagreed with someone, only to go and later drink a few beers them. Just because you disagree with someone doesn't mean you hate or dislike them, and there's no reason to take it personally. It's fine for someone to hold different views than you own.
An example of this are hate-speech laws, which I'm thankful that the US doesn't have. Personally, I consider them horrible mistakes, but I respect that others disagree. FWIW, the reason I disagree with hate-speech laws is that I think you should be able to openly speak your mind on any topic, because it means you can have a discussion and learn from it. If you can't have an open discussion about some topic, you might never be presented with the opportunity to rise above whatever might've lead you to some terrible belief.
I've certainly said a lot of stupid things online, and every time I've been called out on them I think I've grown and learned a bit. I have no doubt I'll continue saying stupid stuff, because in many cases I won't know any better, and I fully hope that others will call me out on it.
Also, while I've been critical of Linus' approach in the past, I think given that his standards are well known and consistent it's probably not that hurtful if he rips you to shreds over a patch because its well known that thats just what hes like.
It's not easy, but it's powerful.
It's absolutely harder. But when you're successful with it, you also start to disarm people like that, and get others to seek you out instead of them.
I think you mean attempt to take advantage of it.
I would assume the author would respond with "Kindly correct them."
I also get tons of free shit by just being nice to service workers.
Contempt is one of the worst regards a person can hold for another--perhaps even worse than hatred. It's a fundamental lack of respect for another's worth, either within a domain or more generally.
One can muster the will to express kindness for someone they dislike. But, it is virtually humanly impossible to be kind towards those one holds in contempt.
Kindness actually triggers the exact opposite of the 3 things above: suddenly everything seems like no big deal and nothing ever changes. Just great: now you’re setting yourself up for several more unpleasant interactions in the future, instead of just fixing something from the beginning.
There are a lot of other considerations too...
For one, the person “yelling” is usually not the only “unkind” person in the interaction, even if that’s the most obvious one. It is unkind, for instance, to be a lazy person who goes into situations utterly unprepared, showing no respect; at that point, YOU aren’t being “nice” so why do you expect niceness in return?
And sometimes niceness gets in the way of well-understood, efficient processes. On a mailing list, say, you’re better off making a direct statement that isn’t wrapped in two extra paragraphs of polite tone for everyone to read through. And heck, when you’re driving, you can create MAJOR traffic problems by being “kind” instead of just following the rules (ironically bubbling back and impacting 50 people for a mile because you wanted to be “kind” to one person; just watch some videos).
1- you can't be kind without appearing weak and
2- being blunt and being kind are two different things
If however you want well deserved respect and kindness, show that you excel at your job, you are able to deliver for me in a timely fashion and exceeding expectation.
You can't handle being criticised ? You have no business being in business, go open a charity bookshop.
One has to understand, developers like in any other creative industry can go off on a tangent by themselves if not given direction explicitly, sometimes that means being very much assertive and firm.
If that is perceived as being unkind then tough luck.
"Your worth is what you do for me" is the kind of transactional inhumanity that makes people put guns in their mouths and pull the trigger. Don't make the world a worse place if you can help it. And you can.
I agree completely.
However, to provide another perspective, that might be more convincing to people who have inclinations towards "what can this person do for me":
You'd be amazed at the difference in results you get through kindness and empathy (while still being direct and clear). You don't just get people who do what you say and respect your technical acumen. You get people who enthusiastically support you with others, who communicate your view because they see it as right, not just as a warning to others who might face your ire otherwise. You get people who actively seek you out as a resource and pull you in on projects, rather than people who do any high-risk-high-reward work behind your back because they don't want you to see it until it's beyond all possible reproach. You become a person whose input is enthusiastically valued, rather than grudgingly required. You build up a cadre of technical engineers and other future technical leaders who value and support the same things you do and understand the same considerations you do, which means the next time a similar issue arises, you might not have to be the person to deal with it personally, because you've mentored those around you.
And while this might depend on your personality, personally, I find that it produces a working culture that gives me energy to take part in, rather than one that saps energy.
So, even if your outlook on life is "how should I behave to get the results I want", I still think you'll find that empathy and kindness will produce better results for you. Leaving aside all the reasons of basic human decency.
If I'm a new developer on your project, I expect to be treated like a valued contributor until my conduct reveals me to be something else.
If you don't like how a team member or contributor is behaving, have that conversation. I totally understand that in the picture you paint, you'd want to be unkind to that person. That makes sense, as they do seem awful, but really that's the manager's fault for: a) hiring them or b) not working to improve their performance or putting them on a PIP, or c) not firing them.
I read the parent comment to be a brusquely phrased statement of that idea.
So, you want to take a developer who is enthusiastic and lacks trepidation, and who is simply inexperienced, and crush that tendency with unkindness?
It's a mentorship opportunity. You recognize the hazard of refactoring working code (especially e.g. without a strong test suite or when there are higher-priority things to be working on), and they don't. So teach them that, and the reasons for it, and then they'll be a better engineer and a more valuable resource for your team. As well as someone who respects your expertise, rather than someone who resents you for snapping at them.
Kindness is a default; beyond ethics or simply being a decent human, it is simply pragmatic to assume good intent on the part of those around you until proven otherwise.
Respect, on the other hand, is earned. Far too many people think they're owed it by virtue of position. To be blunt, fuck that. You can make me fear you that way, but you can't make me respect you.
And a great way to lose my respect is to make everything all about you (are you working on what I think is important?) or demonstrate a lack of empathy. (Which is not to be confused with sympathy.) In fact, showing an inability to empathize might make me pity you - it is a disability - but not want to be around you.
The internet isn't serious business. The vast majority of people don't come close to working on "mission critical software."  At least anecdotally, all of my software managers have been nice and compassionate, if a bit lofty in thinking. If you have a manager that bad, you should leave.
It isn't necessary to be so completely unwilling to give slack on the internet just because it's software development. It's not an elite club. It's just another hobby or another job.
Nobody will die because of a pull request. But people will really feel bad if you slam them for sending it.
Kindness, not contempt, should be the default.
Heh unless you paid me commensurately to warrant this treatment, I would never work like this. As your direct report, you need to _earn_ respect for me. Respect is a two-way street and management doesn't get it simply for being management or having a higher title.
One also has to understand that there ways of achieving this without falling back to command-and-control models of management.
I spent 2.5 years in Pivotal Labs before switching to Cloud R&D. My interpretation of the doctrine I imbibed is that no line of code exists until a test justifies its existence, no test exists unless a story justifies its existence, no story exists unless Product and Design have done some legwork with real users to say yes, this make their lives better.
The system works on trust and empathetic mutual professionalism. And it works very bloody well.
being unkind is also relative, it doesn't necessarily equate to being a jerk or going out of your way to upset someone, no, au contraire, some people perceive unkindness because you dont go around the office shaking people's hand every morning...
i will leave you with one last nugget of wisdom:
Most software nowadays is a non-mission critical where 2-3 average-joe-developer getting paid average salaries will be more than adequate to finish the project. so no , i don't need you to innovate, just need you to execute on a vision as it has been outlined (by a well deserving A player), you do a good job, you become that A player.
I'll say it again for emphasis: the transactional, inhumane culture you are advocating for hurts people and you should stop pissing in our shared pool.
You are not your KPIs and neither are the people over whom you think you lord.
Do it yourself then. Your "vision" is worth jack-shit without execution, so even the "average-joe" devs you're shitting on should be important to you. :)
Or perhaps you're only the "ideas" person...then there are enough overzealous sociopaths in that sphere without you in it as well.
I didn't even notice this nugget of horrible yesterday. You're a "meritocracy" guy, aren't'cha?
Consider this: people who exhibit the attitudes you are exhibiting are very unlikely to actually be "A players" because anyone with options will avoid you for people who don't treat them as machines to bend and break for your personal goals. You can't be an "A player" without the help of the people around you. Why would they help you if you would just consume and discard them?