Hacker News new | past | comments | ask | show | jobs | submit login
No filter: the meanest thing Paul Graham said to a startup (brandonb.cc)
216 points by brandonb on July 23, 2013 | hide | past | favorite | 107 comments



This post is very gladwellesque; presents an anecdote then draws a very general conclusion based on it (being honest is good).

Reading it, it doesn't seem that what made an impression on you was PG's objective criticism of your idea that made your change your mind. It was that he was upset and said something mean. I think that because what you quote of what PG said and therefore must have made the most impact on your decision is his expressions that basically can be translated to "you are fucking stupid": "Moments like these make me glad we invested in sixty-four startups!", "If you want to drive off a cliff, go right ahead.", "like moths for bad ideas."

I suppose you want to be seen as rational actors and would never admit it, but when I read your post it seems like you reacted to PG's anger, frustration, meanness and other emotions, but not to his honesty. The conclusion I would draw from your anecdote is that it is much easier to get your point across if you are emotional, even when dealing with intelligent, technical people.


I wasn't angry. I was resigned. That's why I phrased what I said the way I did. After so many cycles of doing YC, I know a certain percentage of the startups in each batch are doomed. Some people just aren't meant to start startups, but there's no way for them to know that for sure without trying it.

So when a startup seems determined to stay on the wrong path, I won't keep fighting them forever. Eventually I give up. And occasionally when I do I tell them so, because sometimes that wakes them up.


One of the most influential points in my graduate school was when I worked really really hard, and failed to get a result. My advisor gave me this piercing look that without words said firmly, "I'm not angry at you, I'm just disappointed". That woke me up and got me on the right path.


...so what did you change, if your problem wasn't failure to work hard?


technical approaches, mostly, learning not to reinvent the wheel so much, transitioning more towards using well-established techniques in the lab instead of going for broke on a risky project. I think a lot of figuring yourself out is knowing how to properly assess risk (while not becoming risk averse to the point of paralysis), hedge against failure, grab the low hanging fruit, and communicate the above to others.


+1, this is the kind of thing that changes the way you think when you find out.

Sometimes you don't need to be the genious inventing every single part of something to build something great, but we emotionally think we need to, because we often look up to people that did that as it looks too damn hard and wonderful.


Although I learned a LOT in grad school (and in postdoc #1, and in postdoc #2), I will have to say that choice basically ruined any chance I have at getting a faculty position in academia. If you can't join 'em, beat 'em ==> I'm starting my own research institute.


Lots of lessons in life have to be experienced, they can't be taught.


> "... know a certain percentage of the startups in each batch are doomed. Some people just aren't meant to start startups, but there's no way for them to know that for sure without trying it."

People are much stronger and amazing than that.

It's important not to forget that all people have unlimited potential, because we are all the same. However, one factor that cannot be measured, is passion. Passion can come from one or any number of factors - and passion can make the impossible possible.

To that extent, I'd encourage comment readers to grasp that idea you love so greatly, and grasp it with all your heart... for off that cliff lies a river where you can pan for gold.

The other driver's cars were not designed for high altitude nose dives into water. The car designers did not care enough to pay that much attention to detail to handle that dive.

However, for the passionate who can make the finest car, there in that river lies the gold.


Did I time-warp into a Tony Robbins seminar, here?

People have unlimited potential?

We are all the same?

The impossible is possible?

There is a fine line separating rational optimism from potentially dangerous delusion, and your post crosses it.

Stick to science and knowledge. Our potential is finite, we all have very real differences (biological and otherwise), and "the impossible is possible" is an obvious contradiction that renders both terms meaningless, so you're really saying nothing at all.


you toke the words out of my hands :D +1


Wow, you're like the anti-me.

People (with a few exceptions, usually the result of excellent breeding, education or culture) are much weaker and worse than that.

It's important not to forget that all people have limited potential, though some are much more limited than others.

Do you know Plato had a name for the man who was ruled by passion? Every man has to place some principle on the throne of his heart. The true aristocratic man, the philosopher-king, is ruled by wisdom. The timocrat, leader of a warrior state, is ruled by honour.

Going further done the list, the oligarch is ruled by desire for wealth - which at least gives him the motivation to work hard, if nothing else. The democratic man is ruled by freedom, and his life therefore involves being blown around his own whims.

But the lowest of all is the man ruled by passion - the tryant. People here should read The Republic (or knowing HN readers, skim the Wikipedia article describing the 5 different types of polis): he describes modern culture (including Occupy Wall St) pretty well, which just shows that human nature hasn't changed.


too late, but can you quote the exact passage in the Republic? this really sounds cool, that I really don't want to find out if the wikipedia guy and the philosopher phd that came up top on the google and you are the same person.


Wait, hold up.

I don't think his comments translate into "you are stupid." They translate very clearly into "that is a terrible idea." With perhaps a shade of "don't 'pivot' because your feet are cold."

Which is super-valuable guidance!

That's not to say it's pleasant. Clearly communicating that you think someone is making a mistake often sounds like a personal attack. (Maybe that is a personal attack, at least if your decision-making process is deeply intertwined with your identity).

But in person, without preparation, in a few-minutes-long meeting? You're going to hurt someone's feelings. Guaranteed. Just think about how long it would take you to "correct" someone in an email, communicating that they are totally wrong, but not hurting their feelings at all? Heck, we probably each spent about as much time writing these comments as the interview took in its entirety.

There is one personal rule that I feel strongly about, though: never say anything negative in public (or to colleagues) about someone that you haven't already leveled with them about, and ideally given them a chance to change. I don't know if pg did that with the thermostat startup he trashed on Twitter recently, although it's not really any of my business.


The "you are stupid" sentiment comes from this statement: "like moths for bad ideas." To come up with stupid ideas is one thing but to have a natural disposition towards bad ideas is perhaps the definition of stupid. I partly sympathize with the team because it sounds like they were desperate, or at least acting that way. When that happens, rationality fails and your criteria for ideas gets too relaxed and your guiding motivation is to do what's easy and relevant for the most people rather than perhaps providing significant value to a smaller group. The latter is much harder since you are engaging with people you won't be as comfortable with compared to "your mother".


I didn't interpret it as "you are stupid." It's actually surprisingly common for first-time founders, even smart ones, to have terrible startup ideas.

It takes time to develop good judgement. Many ideas that seem initially appealing turn out to be tarpits that ensnare founder after founder. And some ideas that seem crazy at first are actually good. And many ideas are bad initially, but "ripen" as technology changes.

For example, WebVan was a huge failure 12 years ago. But today Instacart is providing a very similar service, but in today's context they can do it without burning tons of capital.

Part of the benefit of YC is that they see so many startups that they can develop keen intuition into which ideas are most likely to work. Not perfect intuition. But definitely better than the average first-time founder.


> I don't think his comments translate into "you are stupid." They translate very clearly into "that is a terrible idea." With perhaps a shade of "don't 'pivot' because your feet are cold."

Consider this: If some random person after listening to you presenting your idea says a. "that is a terrible idea" or b. "you are fucking stupid." Does it matter? B may be more unpleasant to you, but both A and B express the exact same opinion about your idea.

Anyway, I think you missed my point. Nobody cares if feelings get hurt, but expressions like "like moths for bad ideas" aren't brutal honesty.

What would Spock have done when brandonb and his partner presented their (supposedly well-researched) idea bout changing the company? He would have said "My instinct tells me that your idea is detrimental to your business. But this meeting is to short for me to analyze all the facts and formulate a final opinion. Please get back to me next week and I will explain what I think the major weaknesses of your business plan is."

That would have been unfiltered and brutally honest. But it is also possible that it wouldn't have persuaded brandonb to change his mind.


I think we're interested in this because frank feedback is an age-old aspect of 'geek culture', but it's obviously not that important; I'm sure neither of us is too invested in his point of view. Continuing the thread:

I agree with you about people using Steve Jobs-esque "emotional towel-snaps", and how reactions to them may thus be emotional and not rational. Check.

I'm saying, though, that if you clearly (and quickly! within minutes or seconds!) communicate 'you came here for my view, and it's that you're wrong. don't do that', people will read that as emotional, maybe as 'mean', regardless of whether there's any explicit emotional content there. No one likes to hear that answer, so their distaste at getting an answer they don't like makes them view it as an attack.

But I don't see that anything in pg's cited words demands a translation to "you are effing stupid!" It's reasonable for an adult who is basically paying equity to get actionable advice from a trusted investor not to interpret said advice as a personal attack.

To illustrate my point: your 'Spock' answer assumes a much friendlier reality than pg may have seen for them.

What if 'Spock' is convinced they were having a hard time buckling down, that if they changed focus that late in the game, they'd have died?

Then he would have essentially paraphrased pg's response: 'I am as sure as I can be that that is, qualitatively, a terrible course of action. You've nearly died already [moths for bad ideas], and I'm sure you will be dead if you do this. My time's up, so let me know what you decide next week.'


Yeah, I have the same impression.

I prefer Linus style - he at least gives the reasons why something is bad idea, why he won't accept some patches etc.

I don't know about PG, but the blog post portrays him as someone who likes to use argumentum ad verecundiam.

He could at least provide list of some startups which had same/similar idea that failed, for further studies. And maybe he did, if so, it wasn't PG fault that the author of blog post picked irrelevant part, that says nothing more than "you're wrong, because I say so".

For me no-filter style is Linus style, and I really like it.


> I prefer Linus style - he at least gives the reasons why something is bad idea, why he won't accept some patches etc.

Yeah. Basically, when Linus flames somebody, the "mean words" are decoration. They don't actually form the meat of the message, which is almost always technical and detailed. If you had an email filter which deleted all nasty words, you'd still get very useful information from one of Linus's flames.


PG is an authority. If he told me in person that he thought my startup idea sucked, I would listen to him.


I don't take that as a rule, if even after a really nice and long talk about the problem with motives and perspectives analysed I still think he is not right, I would try it my way.


It seems hard to use emotional arguments against logical decisions, and vice versa. Given that their initial decision was emotional ("I want to build something my mom will use!") it seems likely that an emotional counter-argument ("you respect me, and you're losing my respect by acting illogical") was more likely to work.


Agree.

Also would say that if you question somebody in the wrong tone of voice or are vehement enough you can make anyone question a decision they have made.

Here's an example.

You do research and find what you think is the cheapest airfare for a particular destination at a price of $340 USD by checking all the usual suspects. And more.

Someone else who you respect says "now are sure that that is the cheapest price and that a fare of $300 isn't out there?"

You say "yes" but they persist and say "you are absolutely sure right?"

You say yes.

They say "ok will you bet me $500 are you that sure?

All the sudden you are going to 2nd guess and feel that you might have perhaps made a mistake or missed something.

Certainly more than if someone just says "are you sure that's the best price"?. In the case of PG he was daring them and doing what I call "putting a hex" on something by being so vehement that he was right and they were wrong. And that they were going to suffer the consequences.


"Someone else who you respect" ... but is he an expert on flight prices specifically?

Is he a booking-a-flight guru that a majority of world-class flight-bookers go to?

Does he make money from helping you find that $300 ticket?

If he knows that there's a $300 ticket available, and gives me clear guidance on what I need to do to get it (instead of just teasing me with questions), should I ignore him because he was 'putting a hex' on me with his vehemence?


I would suppose not only PG's emotions but also his personality, track record and the OP's general admiration for him, which all add up to his influenciality.

If someone puts all of this in the balance, you listen, and it won't leave you unmoved. Choose wisely whom you grant such power.


> The conclusion I would draw from your anecdote is that it is much easier to get your point across if you are emotional, even when dealing with intelligent, technical people.

You have to command respect, too. If he was just some random suit angrily telling them they were stupid, they might have complied unwillingly.


This post is very gladwellesque; presents an anecdote then draws a very general conclusion based on it (being honest is good).

This is off-topic, but: if "gladwellesque" ever becomes a popular word, I hope this post gets recorded as its first use.


OH: "People interested in being brutally honest often seem more interested in the brutality part"

There's nothing wrong with creating a culture (especially a non-hierarchical culture) built around trust, good intentions, and honest criticism. But honestly, I see a lot more damage in almost every community I am apart of (not just tech communities, art communities and political communities suffer from the same things) from people hitting people over the head in the name of "honesty" than from bad ideas or bad practices being allowed to continue because no one bothered to point them out.


"Brutally" not only refers to the force applied during the conversation but also the lack of escalation in communicating the sentiment. There is a difference between being passive/aggressive or subtle and giving constructive criticism in an unemotional way. The former is easier and ineffective whereas the latter is an exercise in empathy and self-control.


"Brutally" may be one of the poorer choices for a modifier intended to communicate a lack of escalation.

Maybe "flatly honest" would be better?


It's a little simplistic.

Try having 'no filter' in a marriage and most of you would be headed for divorce.

Instead, you need to say things in a manner that your target audience will understand. Sometimes, it takes yelling. Sometimes, you can just be silent and have an even greater impact.

If my kid is throwing a tantrum, I sit down and listen to what he is saying and give him a message in terms that he is likely to understand.

The key is that, if you are dealing with a bunch of headstrong engineers, you might have to push them a lot of get through. If you are dealing with someone who has a bit more sensitivity, this approach can be totally counter-productive.


"Try having 'no filter' in a marriage and most of you would be headed for divorce."

Agree. Most business owners as well would have little success with this unless they have desperate employees where desperate means "can't quit for economic reasons".

Except for celebrities (of who PG is certainly a celebrity with the majority of people who get into YC or are on HN (although not as much)).

Jobs can do this. PG can do this. Others of equivalent stature can because they have what others want. So those "others" will eat all types of shit in order to get their ticket punched.

Most people who aren't desperate for their job (or an investment) will not put up with abusive type behavior. They will walk. For the record I had done that back when I was in college. First time it happened.

Money and tolerance for abuse are closely related.

For the record the thing that bothers me more is the tone and anger as opposed to the words.


It's also worth noting that I believe PG's reputation is mostly even-keeled. The incident described in this post sounds like an outlier. If you're regularly escalating to this level of rhetoric in code reviews or meetings at work, you're probably not using the same criteria for deciding when this kind of talk is justified compared to Mr. Graham. I've seen more virulent language on this site over relatively innocuous comments, which is one example of something I doubt PG would endorse. Who knows if he would even endorse what he said in this conversation, albeit this time it did work.


"The incident described in this post sounds like an outlier."

Would agree that was the impression I had as well.

Strictly for discussion purposes, from my experience, behavior that starts out as an outlier (which works) ends up being the modus operandi if the lack of kickback and inevitable positive reinforcement (in terms of compliance) keep coming.

In the same way that a child not corrected or disciplined will continue acting out especially if they get their way by doing so.


The tone/anger is part of the communication.

But you can be 'angry because you suck' or you can be 'angry because I believe you are better than that'. It's a world of difference.

PG has the decided advantage of having an audience that knows he believes that they are capable of greatness. That makes it a lot easier to say things like that.

I'd say that, for me, respect and tolerance is closely related.

My point is that you need to consider how your audience will hear what you say which is, by definition, a filter. Some of my employees need constant feedback about how they are doing. Some just need a nudge. Knowing the difference is really important.


nothing wrong with pushing back.

its becoming noticeable how americans are turning into babies though


> Try having 'no filter' in a marriage and most of you would be headed for divorce.

Might be a small point, but people marry those who they like. If you change into someone with no filter, then yeah you'll get that divorce. If you've always been that person with no filter, then your marriage is safe considering that's the man your wife/woman your husband married. If expectations get set at a certain place and then suddenly change, friction will happen, regardless. Even if it's going from "no filter" to "tactful response", that's going to cause interpersonal friction as well.

PG seems to have cultivated a perception of "no filter", and so when a person hears, "Let me get Paul", you generally know what's coming. You don't join YC to get patted on the head.


People change over time more often than not.


People get divorced as often as not!


These are qualitatively different kinds of relationships. PG functions as an advisor for YC startups, someone who can influence the direction of your life greatly for a period of time. The difference between failure and success of an early startup can come down to a few critical days or weeks, and living in a version of reality that your ego has warped to protect itself is usually death for a startup. The most important thing in this case is to break that psychological bubble as fast as possible, and harsh words are one way to do it.

One's relationship with a spouse or a child is different. You will generally have many years to influence their thinking, and they won't usually suffer if you do it slowly.


> Try having 'no filter' in a marriage and most of you would be headed for divorce.

Quite a generalization, but just because you are touchy doesn't mean everyone is.


This is nothing compared to the meanest thing a CUSTOMER has said to a startup. Be thankful he doesn't have the power to complain to everyone over twitter, facebook, and the web about your business and the power to sue you if stuff goes wrong. I'll take a harsh-sounding, well-intentioned comment from PG over a truly-mean, ill-intended customer complaint any day. But often even the customer complaints are gems in disguise as well.


Very true. The customers that were the biggest pain in the ass also provide the most accurate, detailed feedback. But it can be hard to hear when you've been up 20 hours working on something. You need to steel yourself and just try to separate out the objective feedback from any apparent rudeness.


> And that has a corrosive effect on culture. Those negative thoughts don't go away, and when team members repress doubts, resentment builds. Passive-aggressive behavior starts to predominate, politics brew, and problems linger on without being solved. So although people may hold their tongue intending to be nice, the result is that a more subtle kind of meanness takes root

This is so absolutely true that its painful, because I have seen this first hand.

The thing you must do is try to figure this out before you join a company, because this tiptoeing around people's feelings seems to work really well at shit and spectacularly average companies. So if you go into a place where this is the accepted way of doing things and try to be the straight shooter, you'll find yourself on everybody's shit list because that's just the way things are done, and in short order (depending on the level of sociopathy that prevails at your particular firm) very strange things will start to happen to your career.

It seems to be that at companies that are doing really amazing things, with really smart and accomplished people, this is not the case.

I also think that this might be a function of culture in the city you work, So while a more unfiltered approach might not raise eyebrows in New York, doing that in Texas could cause problems


For reference: the B2B company Brandon stuck with is Sift Science, which detects fraud via machine learning.

https://siftscience.com/


Why in the hell would they pivot from something like that, to a generic iPhone app idea?


How is this different from what the credit card companies provide? Or do they really not provide that service or only do it on the backend (post transaction)?


(I don't like advertising on these threads, but since somebody asked me directly I'll reply.)

Credit card companies protect consumers, but not merchants. When you call up your bank to report that a transaction was fraudulent, the credit card company reverses the transaction and withdraws the money from the merchant's account. That reversal, called a chargeback, can occur months after the original sale. Merchants can lose thousands, even millions, of dollars per month due to chargebacks.

That means anybody who sells stuff online needs a way to detect fraud on their site in real time, and that's what Sift Science does.

If you run a web site with a payment form, give it a shot: http://siftscience.com.


I appreciate you not overly pitching - I'll check it out. Separately, why aren't you mentioned on the team page? Are you still involved?

https://siftscience.com/about


It looks like it might be different from checking for monetary fraud. It almost seems like it monitors activity on your website to determine any abnormal activity?

Maybe it's more along the line of loss prevention positions at a retail store, which monitor customers' behavior in-store, and follow the ones that look suspicious. Applied to web activity, it seems like it might yield interesting results...


> Actively fostering a culture of "no filter" is painful at first. But like exercising, the more you do it, the easier it gets. And it's better than the alternative—death.

No.

You say that people tried to tell you what was wrong, and that you didn't listen.

Better communication doesn't mean "start with being blunt", just as better communication doesn't mean "dance around the issues but never talk directly about them for fear of offending people."

Or, let's see how this works:

You're an idiot if the lesson you get from this is that 'no filters' are a good idea. Only a dumb-ass would reach that conclusion. The same kind of fuckwit who would ignore all the smart people trying to give them information and who only gets the point when someone is exasperated enough to fall into rudeness.


I see your point, but different sides of the communication need to hear different things to make things better. Whatever side you're on, you can only change what you do, not what other people do. He's talking to the "tellers" in this case, based on his experience as a recipient, telling them what works.

I totally agree with you on the stupid false dichotomy that comes up in all these discussions. It's difficult but possible to get your point across efficiently without being rude. If people think you're being rude, well, sometimes that's a listener problem and there's nothing you can do about it.


"Crocker's Rules" (http://en.wikipedia.org/wiki/Lee_Daniel_Crocker) seems relevant.


It's possible to be honest and tactful at the same time, but doing so requires developing quality real-time filters, which is hard and takes lots of practice. One hypothesis is that powerful people have the luxury of not developing such filters because they can better get away with alienating the targets of their (often, to be fair, dead-on) criticism. Mistaking correlation for causation, people then conclude that operating without such filters is the cause of such power, rather than an effect.

As a thought experiment, consider the Jobs quote from the OP:

We are brutally honest with each other, and anyone can tell me they think I am full of shit and I can tell them the same.

Now, what would you guess was the rough ratio of brutal honesty directed to Jobs to the brutal honesty coming from Jobs? In physics parlance, I'm guessing it was << 1. Jobs was human, and even an advocate of brutal honesty would probably rather hear "Steve, I need to talk to you about a delicate subject..." than "Steve, you smell like shit and need to take a shower." [1] Considering that Steve could fire you at will, which tack would you take?

[1] I once had the pleasure of hearing Mike Scott tell me and a group of other Caltech grad students about his first duty as Apple's first president: tell Steve that, contrary to his belief, a "fruitarian" diet does not result in zero body odor. If memory serves, Mike was wisely oblique and tactful when bringing up the subject.


Personally, I would much rather someone be brutally honest with me, rather than beating around the bush or talking in innuendos hoping that I will figure it out. PG is an expert in his field, and there is no point for him to waste either parties time participating in this charade. Being blunt likely accelerates the learning curve by gutting the conversation of red tape. Getting to the meat and potatoes of the conversation.

p.s. This is also why I respect Linus.


I genuinely do not understand why every discussion about this has people presenting the same two extremes - "Fuck off you idiot" on one end and an inability to give any useful information on the other.

In between those two extremes are a wide range of responses that can express anger at having your time wasted, through gratitude for something but inability to use it because of $REASON, through all sort of other stuff.


No one's defending the "fuck off you idiot" extreme, I don't believe. Linus, PG, no one successful gives that kind of criticism very often. It's usually, "fuck off for reasons x, y, and z. You're acting like an idiot."


Going through life being brutally honest would be a hell of a way to live. There obviously needs to be balance and there is a time and a place for everything. I just think that being brutally honest is the time and place when you are asking an expert for their opinion, like seeing a Dr. about a medical issue, he is not mean, but will tell you directly the facts, and his professional opinion on how to help your situation.

I guess brutally honest is open to interpretation too. I do mean, honest, professional, constructive feedback, but don't blow smoke up my ass when I need your help, because that is not what I'm there for.


Often because someone accused of being too harsh presents the polar opposite as being the only other option.


I've worked in companies where it ends up a trainwreck, and one big reason is that everyone gets excited over some project, and keeps pivoting and not concentrating on bringing a project to an actual product for positive growth.

I respect PG, and the simple fact his time is limited, and his terse replies are usually to the point and justified.


Paul can be terse, and sometimes you don't want to hear what he has to say, and he can be wrong; but he's never a bully.


sounds to me like he was using his position to bully these guys away from pivoting. If he hadn't been an investor his opinion, like those of their other colleagues, would have gone unheeded. How is this not bullying? Personally I don't even think it was a mean thing to say, it was just honesty, but the whole drama stuff of throwing up arms in exasperation rather than just professionally explaining why they were about to make am mistake is just bad manners.


There are several factors here and I haven't seen anybody post them:

-PG has surely seen lots of mistakes like the op was going to do.

-He has seen how they end.

-he has tried persuading those founders in several ways, and probably failed too many times to be satisfied.

-probably (I don't know PG beyond his essays and concerences) he knew that giving polite advice was the best way to keep the OP on his track to failure.

-so be did the equivalent of a Doctor starting a CPR: a heavy punch to the chest. That may seem a rude behavior to a external observer, but in fact he is saving a live.

I suppose that his mean comment was studied and acted, not REALLY emotional. But it worked none the less, it doesn't matter what other people think about that, his role is to keep startups growing, not external people happy about his kindness. PD. CPR (CRP in spanish) and the enter button doesn´t seem to work on the Iphone.


I had a professor at Stanford who had a policy to always give positive advice, unless you said "give it to me hot", in which case he gave you the truth. I always asked for advice this way and it saved my butt a few times. More people should have this "brutal honesty mode".


Our second week in Y Combinator, we almost pivoted away from building a B2B startup

I thought a "pivot" was a change in target market in an attempt to find product-market fit, OR a change to a whole new idea in response to a failure to find product-market fit for the current idea. The lean startup book defines a pivot as a "structured course correction designed to test a new fundamental hypothesis about the product, strategy, and engine of growth.".

From that little blurb this sounds more like a "I'm bored with this B2B thing, let's do something totally different and B2C" moment rather than an actual pivot. Or am I missing something?


Personally I believe that no filter communication works best in conveying ideas clearly and directly. However, it only work if there's established trust and respect in the relationship. In addition, no filter doesn't mean be abrasive (like Linus). I've found that with some pretty competent people they can shutdown in those situations. Clear, honest, no fluff communication is essential, but it's also essential to keep the conversation about the work and not about the person.

A combination of appropriate encouragement, honest criticism, and healthy respect is needed to bring out the best in people.


Great post, Brandon. Getting object & honest thoughts, opinions & feedback is vitally important to both teams and each one of us. To help our team be more open, I've tried to use regular retrospectives where we highlight aspects of projects/sprints/teams that went well, poorly, and some that were both. I've also tried having time away from work, but found that without the focus on feedback, people are reluctant to offer criticism.

What practices have others used to foster critical feedback? What types of personalities are best at bringing this out of others?


If you want to be an effective communicator, it may be a good idea to speak to different people differently. Brutal honesty works for some people. Other people will just get so upset or defensive that they'll stop thinking or choose not to associate with you.

I'd suggest that when it comes to manners and you don't know the other party that well, you might benefit from playing your cards as conservatively as you can given the time you have available. You can get ... blunter ... a lot more easily than you can take back hurting someone's feelings.

> Actively fostering a culture of "no filter" is painful at first. But like exercising, the more you do it, the easier it gets. And it's better than the alternative — death.

Supposedly, old Chinese courts used to communicate with the Emperor via the interpretation of heavenly portents. I seem to remember that the fool in some European courts was meant to criticise the ruler too.

Whether true or not, however, in either of those cases you'd be abstracting the filter out into a procedure and avoiding retaliation from the ruler.

If you think that the problem is that people have their egos invested in particular solutions, then I suppose that you could take approaches liable to lessen attachment in the generation stage too.

Anyway, I'm not really sure that the only alternative to no filter is death, is what I'm saying ^^; (Though it seems plausible that the alternative to no honesty is.)


This is great. I once was a part of a startup. I had my doubts about the idea but the other co-founder assured me that the market would work well. I was the tech guy and had nothing else to do at that point(and also coz I love to code) played along.

We talked to a lot of people about the idea and they all had "kinda" nice response, too much filtered. The startup didn't work out. Those were the times when I wished the comments were as unfiltered as the one made by PG here.


It's a good question to ask yourself all the time, not only when you are working on a startup. "Am I failing? How?" The more you ask it, the more evaluate your own progress according to whatever goals you have, the better you get at managing your own progress towards goals -- and the less you need other people to manage you.

That said, kudos to pg and team for being honest and to the point.


Good post until the Linus reference without any context.


Heh... I laughed. I got the context. I suspect you did too.

As a matter of fact as soon as I saw "SHUT THE FUCK UP" I knew that when I moused over[1], I was going to be looking at a link to the LKML.

Ed: [1] And I haven't ever read the original thread.


You probably should read the original thread. People like you are exactly why the reference is out of context, you don't understand how egregious the mistake the person who Linus is replying to made.


You are mistaken. I have read and replied to people like the people you accuse me of being, and defended the right and necessity of the newsgroup flame. Some people fuck up just that bad. You can't even sugar coat it.

It's people like you who...

Just kidding!


Do you think it's bad because you disagree with the point of the Linus reference, or because you didn't understand it without context?


You imply that Linus was overdoing it, he wasn't.

Mauro wasn't asking for feedback, he acted in a way that resulted in strong negative feedback, that's why it's not really relevant to the rest of your post.


I disagree with the Linus reference being brought up without the surrounding details. It's too easy to just point out someone who is using foul language.


Minor distracting nitpick: the post title is misleading. I thought PG was the one who said "No filter." I realize the title has no quotes, but missing quotes seems like a smaller mistake than requiring the reader to decipher what it's supposed to mean ("PG has no filter against speaking his mind, which led him to saying the meanest thing he ever said to a startup").

Better title: "Paul Graham dares to criticize, and why you should to." Or just drop the "No filter:".

Also, for the first half of the article, I thought "no filter" referred to the founders who switched focus on a whim, ie "no filter against other alluring ideas."


This was a great post. Honesty/Transparency/Bluntness is just critical. Related to this, it requires a culture where people don't have an emotional attachment to their ideas or work (at least during the development stage).


We shouldn't aim for "brutally honest", we should aim for "honest" (meaning telling the truth, but trying our best not to hurt the other party's feelings).

Also, the author seems to assume that the person being brutally honest is always right in his/her criticism. But what if he/she isn't? This is the situation many employees find themselves in: Their boss lacks the qualifications to bring valuable input to the table, but still thinks he/she should be allowed to be "brutally honest" (which in those cases basically means being dumb and a loudmouth at the same time).


I enjoy reading HN very much, Most start ups that HN invests in are started by a group of very young people and from what I understand, the amount of money invested is relatively small, which is probably wise. But as someone who has, let's say been around awhile, let me just say that it probably would have been really helpful to have a resource like YC when I was making all my mistakes (ahem, several fails, leave it at that). I agree, PG sounds like he knows when to shake you up a little, I wish I had had a little guidance like that back when. YC'ers, pay attention.


Off topic, but Linus is my hero after reading that e-mail.

After that story which uncovered sheer idiocy of glibc developers who break user code between two minor releases. Was it Drepper who broke memmove?


Was this recent? If not, how did the advice turn out?


This was two years ago, and so far PG's advice is turning out well: we were able to raise a seed round, series A, get a bunch of customers, and hire an awesome team.

You can never be exactly sure where the other road would take you, but one of our employees, who was very encouraging about the local events idea, told me after he decided to join: "There's no way I would have joined if you were working on local events. A user, maybe, but joining you, no."

And in the same period of time, I've seen a lot of startups focus on some variant of local events, and most have pivoted or failed completely.


Local events is one of those ideas PG says he outright hates. I remember being surprised when you guys said you were working on it during YC and happy when I heard you moved into fraud detection.


Brandond, would you care to give insight why the local event idea is bad ?


What's with the double standard? There seems to be a disconnect between what people agree can be said to someone's face and what people can say on a public internet forum.

I see defense of language in the article and that doesn't add up with what people refer to as 'fatuous vitriol' and are up in arms about on HN comments.

I guess there are no hellbannings in 'real life.'


But whose honesty are we talking about? I agree with the heart of this, but very often leaders being "brutally honest" to the point of being rude and inconsiderate have the effect of causing others to become too afraid to speak their minds, lest they trigger an angry response.


TIL: Paul Graham is one of the nicest people on earth if that's the meanest thing he could say.


Never compromise on clarity.

Nothing wrong with the proper application of a filter. Filter out words that don't improve the message, or criticism that isn't constructive. The problem with your filter is that you had it tuned to filter out anything that disagreed with you.


It is so hard to get some people to understand that criticism is not always meant to be an insult.


Just curious: what was the idea of the mobile app for local events?


I'm interested in that, too.

I go out a lot (in my home town, and even more so in holidays), and I'd love to have an app to give me suggestions of events (special nights, museums, openings, and such).

Why was it a bad idea ? Have brandond and al. research the market ?


Having worked on a local events app in the past (which I ended up abandoning), I have a few thoughts. These aren't necessarily unsurmountable, but they were sticky points:

- Event database curation doesn't scale very well. If you want high quality and accurate content, you're either going to need to have people double-checking stuff or come up with some nice algorithms to pre-filter it.

- Event entry is a hassle. At the start, you're going to be entering a LOT of the content by hand. This is pretty normal across a lot of different types of product, but with event data, the data has a very definite shelf-life. As soon as the event is over, you've lost a piece of data and you're going to need to replace it with something else.

- Chicken & egg: when you haven't grown to be the place to search for events in a city, you're not going to have people entering events for the city (e.g. people who work at venues who want to advertise their events). There are so many different apps that advertise events for you, and putting events into all of them is going to pretty onerous.

- Facebook: for a venue, or a band, Facebook is a super easy captive audience. I follow venues I like on FB, and they advertise all of their coming events right there. That's your competition.

- Reasonably high cost to drive traffic. If I recall, we were paying somewhere around $0.25/user that came to our site. We got decent inbound traffic, but almost no sign-ups, and even fewer self-posted events.

- How do you monetize it? Your two big competitors are Facebook (free for posting events) and posters stapled to telephone poles (25¢/poster to print it).


Thanks for your feedback. This is very interesting.

> Your two big competitors are Facebook (free for posting events)

Facebook is not exactly a competitor because it's a walled garden. So you have to be part of the right groups or be suscribed to official pages to get events.

It don't know about you, but all this facebook overhead is starting to bore me. I love salsa, hip places and fancy. I want the best events for me NOW. I love dive bars, cheap beer and sweaty rock. I want the best events NOW.

> - Chicken & egg: If you can't populate your event base with pre existing data, you may have a chicken/egg problem.

Unless you have some magic crawler technology...

I'd love to have a access to a event central database that curates events for me matching my taste and previous choices.

I could pay a flat fee for a mobile app that could do that. And I wouldn't mind it having ads for other places/events that could suit my taste.

I have a hard time imagining that events curating is not a real problem. Finding new places and fun events to go to is actually a pain.

Is it a monetizable pain ?

The linked article implies that Paul Graham thinks that this is not worthy of a shot (or a second shot). Is it not monetizable enough ? Is it a nut too tough to crack ?

I wonder what the reasons are (I wish I have more research on the subject).

Well, this problem got me thinking.

If you want to keep talking about it, here's my email name.is.carl (@) gmail (you know what)


Thanks for writing these tips...

So hard to resist making a nice event finding app!


There's an app called Now that fits that kind of description. http://www.getnowapp.com/


"I want to design something my mom can use".

Ahhh, remind me again where I have heard this before. Thank god, pg shot the idea down.


There is saying that people will always say nice things. You need good friends or pay small fortune to hear bad things.


So what was the terrible B2C idea?


From the OP: "So we pivoted to a mobile app for local events."


Yep. That's it.


Not a good sign if you pivot twice before you even get to the top of the hill.


Heh, good for them. I know so many companies in local events it's silly.


American society is way too polite and "sensitive" now days. Go read Mencken and Twain, and editorials in general from 90 years ago. You couldn't write invective like that anymore; people would freak out and demand apologies. It's sad.

Seems to me the most effective organizations don't pussy foot around stark language. Read about Bill Gates back in the heyday calling people idiots and yelling at them.


My thought on this: a "no filter" policy is good, but it requires sensitivity. Paul Graham can shoot down a bad idea (that's a service, not a slight) but not a person, which is an important skill because good people come up with bad ideas all the time and someone has to do the culling. Few people have that ability.

Paul Graham is not (from the sounds of it) an asshole. Constructive criticism is good, but assholes are toxic. They shoot down bad ideas, but also good ones, and generally make people insecure. They aren't taken seriously after a while, but they inject random, highly visible, embarrassment that can effectively target anyone and it has a chilling effect.

People create and refine these filters and then go too far in the other direction. They know they aren't assholes, but they don't even want to look like assholes.

Additionally, most people spend most of their organizational lives as subordinates, in positions where there is literally no upside (and considerable downside) to being a bearer of bad news.




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

Search: