Au Revoir | Hacker News Reader

Au Revoir

blog.docker.com

898 pointszapita287 comments

I remember being introduced to Solomon in Dec 2008 (a friend of a friend) and he came to my apartment in Los Angeles a few months later to demo his "dotCloud" project. He told me how he was using Mercurial to have a version-controlled repository of an entire root filesystem (including char and block device special files under /dev; he had to extend Mercurial to add support for them). I found this pretty cool and I could see the potential. He could check out multiple copies, make different changes, merge them, etc. It was basically a revision-controlled VM/container/system image. I was working on somewhat related technology at the time, http://qemudo.sourceforge.net/?, a web-interface to QEMU virtual machines where whole VMs could be created in <1 second by using copy-on-write disk images (QCOW.) Everyone I showed this to back in 2006-2008 was dumbfounded. The idea that VM creation could be a lightweight operation was brand new back then. Normally it involved copying a multi-gigabyte disk image. All the pieces of the tech stack were available to do this, but no well-polished product that combined them together existed.

Of course dotCloud ended up being a precursor of Docker itself, and took the idea of "lightweight easy-to-create container" to an even higher level. shykes if you are reading this, hi!

I love the art made for the post. It's so emotional and powerful. The letter told the story, but the image made me feel the story. So simple, but yet so powerful.

Also, @solomonstre has changed his Twitter profile picture. This is brilliant communication. He lets everyone know that he has moved on to other things, just from his profile picture. I think this is very interesting and clever.

Docker hired a famous french comic book artist for all their graphics: https://twitter.com/laurelcomics

It really brought a beautiful graphical identity to their branding. Very well done!

I really enjoyed this long comic by Laurel https://commeconvenu.com/. It's available online, but only in French I'm afraid... It's the story of how she came to the US for a startup with her boyfriend and how they found themselves in a more and more difficult situation with their cofounders. It's gripping, hilarious at times, and the art is outstanding.

Wow, thanks for the link. That was a very interesting read & sad story.

Yeah, came here to say this https://i0.wp.com/blog.docker.com/wp-content/uploads/solomon... is awesome

Honestly, I was not aware you could convey so much emotion in a simple cartoon. This is just outstanding work.

His little boat seems safer... it's a joke, pls don't read too much into this.

Except that the rigging makes absolutely no sense.

Good - it's not just me!

Or me ... I thought that was a roller-furling jib on the back-stay!

Haha, that was the strangeness: The boat is the opposite way compared to the rig..

I have to agree with you - and has made an excellent transition to whatever is next and built a very strong reputation in the process, and shown enough pride in his product to help let it grow further, even without him at the helm.

It always amazes me how you can convince a founder/CEO is not right man for the job, and slowly push him away.

Companies that succeed are driven by burning vision, not by day-to-day operation people. It might make feel the companies run smoother for some time, but day-to-day executives are harmful to a company core vision leading to boring companies if they manage to sustain profitability in the long run at all.

That's a pretty harsh generalisation.

Edit: I should probably expand on that. Yes, in some cases keeping the founder as a CEO can be important for the morale of the team, but sometimes bringing in outside experience (the so called 'adult supervision') can be very important for a company. Eric Schmidt did a great job doing this for Google, before he assumed the role as chairman, leaving the CEO post for the founders at a later point.

Whereas effective leaders are required to manage not only the day-to-day, but turning the core vision into an actual business plan that can pay employees rather than just burn VC money. My experience is the opposite; I've worked at two companies where the founder has returned to be CEO and in both cases they were highly ineffective.

I'd concur, the generalization of one being better than the other just doesn't fit. I've worked for companies that fit into both situations and it really comes down to whether the founder had the skills to begin with or whether they were willing to admit that they needed help on the day-to-day operations & maintenance, etc.

Agreed. There is this concept of the wheel of energy, people tend to be good at different parts of the wheel:

Starting Growing Containing Completing

Founder types tend to be amazing at Starting, and sometimes at Growing. But at a certain point, you need to bring in the people who are better at the Growing/Containing phase. IMHO most Founder/CEOs don't have this distinction and can't see how what makes them terrific at the beginning hobbles the company as it tries to become a stable business.

i guess it's a cycle. When a company needs to innovate to build its first iterations then the founder role is crucial, but then come the time of finding the best way to make money out of the product, and have a sustainable and well organized company. In that moment, (technical) founders find themselves a bit bored. The ideal would be for them to get back to pure R&D for a year or two, to discover more paths for the company's future. The problem is that carriers in a company only go one direction. Once you are a C-something the only way left is out.

Agreed. Most activities in big companies are not even remotely related to the core technology or idea that brought the company into existence. As company grows, training departments grow to train new clients, technical writing teams grow to write documentation in different languages, technical teams grow to support the many different operating systems and markets, HR department grow to support the many people now working at the company and soon the effort required to coordinate all these people is more than the effort put into the original product. Some founders adapt to the changing needs of the company some don't like all the admin.

There is a reason why most startups don't go big - burning vision is not even close to being the right formula for success. Running a company of 10 people is VERY different from running a company of 100, which is VERY VERY different from running a company of 1000. Just like precious few individual contributors are comfortable with and able to contribute meaningfully in this wide range of environments, it doesn't make sense that a CEO is somehow exempted from that. The day-to-day operations are in fact incredibly different and present unique challenges.

Note: I'm not saying this is what's happening here, just responding to the general comment above.

> Running a company of 10 people is VERY different from running a company of 100, which is VERY VERY different from running a company of 1000

IMO thats a classist lie that keeps people in their place. Any reasonably smart CEO can learn his way from 10 - 10000. Yes, (s)he may not be capable of 10k on day one, but they can grow along the way.

Ehh I'm not so sure. There are different levels of abstractions involved in a 10 person company vs a 1000 person company.

At the 5000 employee level, you are probably dealing with a ton of abstractions that were filtered multiple levels up.

Let's all admit that we don't know because we're not/have not been in that scenario rather than picking arbitrary milestones and making guesses

It's probably a little bit like a Cessna pilot trying to land a 747. I'm sure the plane can almost land itself, and ground control can talk you through it, but wouldn't you rather have someone who has done it before?

Eh, most 747 pilots have probably flown a number of smaller planes before.

Having a number of hours in a small plane in no way disqualifies you to fly bigger planes.

Also in this case they aren't only the pilot but also the constructor of the "plane".

(That said while most pilots started small not all Cessna pilots should go on to fly passengers or even cargo.)

It's your closing parenthetical that's the important point here.

> (That said while most pilots started small not all Cessna pilots should go on to fly passengers or even cargo.)

Unless I’m mistaken, that was actually the point being made; you made it again for him.

Your argument was going the other direction.

No way is it a lie.

The kind of people who enjoy running a company with a ragtag group of 10 employees in the early days are rarely the kind of people who will be able to effectively run a 10000 employee operation.

If you’re growing fast there is no time for on-the-job training as CEO. You need to be effective immediately or else you are only hurting your company’s growth.

Oh my God, the company's growth? Then we MUST axe the CEO. He's just a human, after all. Don't want humans to hurt company's growth. Company's growth. Growth.

When you take outside money then growth or profit become the only metrics that matter. That's how capitalism works.

Exactly, it’s better to let an inexperienced CEO let the company burn through funds and shut down than to bring in someone experienced with managing large teams and implementing a strategy towards a vision. It’s a learning experiment.

If you're not growing you're dying.

...but, if you're growing too much, you might be cancer.

Yeah, the problem is those humans have nowhere to grow. Would you be happy with the same salary, job title, etc for a decade?

When that human and their company has taken tens of millions or hundreds of millions of dollars of investors' money, yes, growth is everything. Investments need to be paid back with interest. Without a lot of growth, investors are just giving away money.

What do you think an investment is? How do you think business works?

The guy didn't get pushed out in any event.

It's funny, because I absolutely agree with your assessment. But then I started a couple of thought experiments.

How about the reverse case? Would, say, Meg Whitman or Satya Nadella, for example, be able to act as a CEO or co-founder of a small company, say < 10 employees?

But then I thought about Bill Joy, Bill Gates, Larry Ellison, Steve Jobs, Jeff Bezos, Mark Zuckerberg, and, well, you hopefully get my point.

Which is, even though I agree that most startup founders probably would not be able to head the same company as it grew by multiple orders of magnitude, there are enough famous outliers that we know it can be done and can be done spectacularly well.

There must be a statistical name for this apparent paradox. (Or am I just experiencing confirmation bias?)

EDIT: grammar nits

Meg Whitman ran eBay when it was 30 people.

You're thinking of survivorship bias

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

It's entirely possible that some big CEOs can't do well with a small company of 10 employees or so. A lot of the skills they have are mostly focused on managing larger organized teams, which is not the same skillset for small teams. It's strategy vs tactics.

Maybe at one point they could, early in their career, but by now those skills may have atrophied.

I think the phrase you're looking for is "exceptions that prove the rule". For every CEO that took a company from nothing to 10k employees, there are so many more examples of the opposite that you can't even begin to name them. You're able to name the exceptions because there are so few exceptions.

But it is worthy of note that right now, the top 6 largest companies (by market cap) in the US are or were helmed by their founders well into their fortune 500 status with many thousands of employees.

It's also worthy of note that these companies are extreme statistical outliers. Not every company is fortunate enough to have someone like Steve Jobs or Bill Gates leading them.

It's also worth noting that all these companies have experienced executive leaders who came on during the transition from 100 to 1000 people. They succeeded with a massive amount of support from experienced top executives. This seems to be the optimal approach, hire in top execs as COO or CEO (then later chairmen or other top advicer) while keeping the founders in the leadership team.

That phrase is widely misused. The intended usage is where a rule is implied by the explicit stating of an exception. For example, a sign labeled "No Parking 9AM - 7PM" states an exception, and implies that parking is allowed outside of the time. The stating of the exception proves the existence of the more general rule.

In this case, the appropriate term would be "survivorship bias". Those who survive are noticed and used to generalize to the group, even if their survival is atypical of the group as a whole.

According to Wikipedia, there are multiple ways to use the phrase. The way I used it is listed under "loose rhetorical sense".

The word "prove" has largely changed its meaning, it doesn't mean "test" anymore, although a few usages from previous centuries survive, such as military "proving grounds." So yes, it's very interesting that an old phrase that once meant 'a circumstance that challenges or makes one doubt a rule', gave birth to the modern phrase that means pretty much the opposite: "makes one more confident of the rule, in the end, after all." But the latter phrase exists, the former is part of history.

Language changes constantly, and meanings often flip from one century to another. The previous meaning of "prove" doesn't obviate the modern usage inside or outside of the phrase in question or turn the clock back. That particular horse escaped the barn, long, long ago along with tens of thousands of other changes in the last few centuries. "Probable" existed before probability theory did, and it didn't refer to odds or math; it just meant "respectable."

I could follow your own suit, I suppose, and compliment you by calling your comment "smart", since that word used to mean "painful" in olde english and you prefer the old ways. But I won't. (I'll just say your comment was interesting, and not quite the full story.) I roll with the common currency: language as it is. Print greatly slowed these sorts of changes, and the internet will slow them even more, of course, as the decades and centuries roll on.

Most idioms I am fine with shifting, or being nonsensical when broken by time. I couldn't care less if people use "I could care less", as there is no mistaken meaning. My problem with the current use of "exception proves the rule" is that it is factually untrue. Not only that, but unlike other factually untrue phrases such as "You catch more flies with honey than with vinegar.", its current meaning promotes bad science. Exceptions are evidence against the rule, unless there is reason to discount the exceptions.

In general, I try to correct it in order to give the benefit of the doubt. Either somebody is misusing a phrase, which is a charitable interpretation, or somebody has a fundamentally flawed idea of how evidence works, which is rather uncharitable.

I thought I explained my usage pretty well. You should go back and re-read it, and also take a look at the Wikipedia page I reference, where it's explained in more detail.

Basically, you can name the exceptions off the top of your head, but you could never do that with the non-exceptions because they're so numerous. Kind of like how in your parking example, you don't need to name the hours that parking is allowed, because it's a rule. It's always allowed, with some exceptions. You only need to point out the exceptions.

I would take your charitable interpretation and say you're misunderstanding the phrase. Since there are examples of this usage all over the Internet and a very long Wikipedia page detailing the various forms, I hope we can move past this wildly off-topic conversation.

After working through a number of startups that grow from 10s to hundreds I don't think it is untrue.

The argument is not what an individual is capable of learning and doing but personality fit. The founder who has a burning desire to unleash their idea to the world is generally not the same person who wants to grow the company at a larger scale.

1) I think the founder treats the company too much like a baby and sometimes the hard realities of the business are ignored. It is nice to have a fresh breath of air from leadership who have not been with the company from day 1. They have a different perspective that I believe can make a huge difference.

2) It is a different type of work. Perhaps the person can make the transition but I don't think that is normally the case. The challenges are different at size 500 compared to what t hey were at size 10 and I think a lot of CEOs who strive in the small scrappy environment do not really enjoy it at size 500. Sure they may trick themselves in thinking it is for them but I think it does more harm than good but it is also a learning experience for a lot of first time founders.

Don't confuse me. I am not saying its impossible to successful founder/CEO at size 500+ but I don't think the type of person who founds a company really enjoys it.

What on earth makes it a “classist lie”?

The implication is that the (especially American) upper class is trenched on this belief that CEOs of large companies require a certain pedigree that may not be based on observable, empirical evidence but rather on a mythos to keep the upper class relatively "pure" and the number of rich CEOs controlled to (mostly) existing class members, while attempting to appear egalitarian ("anyone" can go to an ivy league school; "anyone" can fail upward; "anyone" can develop the skills to be a proper CEO, it's just "hard work" that a silver spoon makes "easier").

The relative rarity of CEO jobs at sizes deemed large helps ensure that empiricism is complicated because the sample sizes remain small, so what the truth here is, who can say? It's tough to science.

It's not classist to want to hire someone with experience for a job that demands it. If your money and your startup was on the line, would you want to hire only junior programmers and hope it works out? No, time is of the essence and you have investors who put their faith in you, so you're much more likely to put people with experience in charge.

No one is questioning the desire to hire people with experience. That's a good desire, usually.

The "classist" question is why is the pool of people with such experience so small and seem so exclusionary?

One specific thing mentioned up thread is the perception that not a lot of small, startup CEOs "graduate" to large, corporate CEOs and asked the question why doesn't "small, startup CEO" count for more experience in these searches than it seems to? The people that believe that classism may be to blame wonder if it is simply and inevitably because they are outsiders to the "proper" upper class or other such cliques. Anecdotally, there seem to be counter-examples on both sides of that hypothesis, and I don't think either side has enough data to prove or disprove the hypothesis.

What's interesting about what you said is that if one were to look at the CEOs of the Fortune 500, this idea of classism breaks down.

CEO of IBM: Ginni Rometty -- grew up in a single parent home, with a mom that worked multiple jobs. Ginni got in to Northwestern and studied computer science and electrical engineering. She certainly didn't have any upper-class advantages. She went to public school.

Geisha Williams, CEO of PG&E (part of the Fortune 500) -- daughter of Cuban refugees -- literally arrived in the US with nothing, her father was a political prisoner in Cuba before being released. He worked multiple jobs to support his family. Geisha started her career as a residential power auditor and worked her way up to CEO. Hardly a "silver spoon" story.

Rex Tillerson (former) CEO of Exxon -- one of the largest companies in the world. Picked cotton and worked as a janitor before getting accepted to the University of Texas with a band scholarship. He joined Exxon as a production engineer and then worked his way up to CEO over the next 30 years. Hardly an "elite."

Tim Cook -- son of a dockworker and a pharmacy worker. Went to Auburn University (hardly an Ivy League, "elite" school.) He worked at IBM from 12 years working his way up to the director of North American fulfillment while earning a Duke MBA at essentially night school.

I could tell these stories all day. There certainly are some elites who do go on to be CEOs, but there are even more CEOs and self-made millionaires that aren't Boston Brahmins or cousins of the Kennedy family.

This whole "classism" idea is promoted by three types of people:

a. those who are actually ignorant about American business and its leaders. Spend some time actually learning the backgrounds of American business leaders and you'll see more "State U" than "Harvard U."

b. those from other countries where a public school girl from a broken home would never become CEO of IBM or a redneck janitor on a marching band scholarship could never become CEO of Exxon -- these are often Europeans where, in places such as France -- if you aren't a male who went to a Grande Ecole, you will never be CEO of a major French company. CEO of Groupe PSA (France's largest car maker) -- Carlos Tavares, graduate of Ecole Centrale Paris. CEO of Renault? Also a Grande Ecole graduate. Former CEO of Airbus? Louis Gallois, also from a Grande Ecole. CEO of LVMH? Bernard Arnault, another Grande Ecole graduate. The "egalitarian" French have nothing but elites leading their top companies. In the US, the most valuable company's CEO? Tim Cook -- Auburn University -- a state school. Rex Tillerson? University of Texas. CEO of Walmart (the top company on the US Fortune 500 by revenue) -- University of Arkansas and University of Tulsa. CEO of United Health? Dave Wichmannm, Illinois State University with an Accounting degree -- not even an MBA and he's the CEO of the number 6 company in the United States. CEO of United Airlines? Oscar Munoz, the oldest of 9 kids in a Mexican-American family in California and the first to graduate college in his family. Some elite. What "classism" helped Munoz? He's the CEO of the world's largest airline.

c. the type of person that is angry at their own life, their own bad decisions or their own lack of success that they want to create a political narrative that supports the "evil capitalism" trope. This is the type of person that feels like they have been wronged by life and society and rather than looking at their own failings, instead attempt to blame some boogeyman. Maybe their daddy didn't love them enough. I don't know. But these are negative people that claim conspiracy when there is none. They themselves just don't think they're good enough and they need an excuse to justify it.

Now, if we were talking about politics -- THEN I might agree that classism has some influence. When we have a brood of Bushes, a variety of Clintons and an infestation of Kennedys, then it's pretty clear classism does have a real impact. But in the C-Suite -- I would argue that the evidence suggests that classism isn't a factor at all, instead it's about hard work, tenacity, luck and intelligence.

This is such an awesome comment. I just wanted to say, thank you for posting this.

The appeal to marxist ideals is deliciously ironic in maerF0x0's comment.

I would like to understand this more, could you expand?

Do you not see the irony?

No, but I believe you could make it clear.

There's irony in complaining about unfair standards applied to CEOs, whose primary aim is essentially to extract the maximum possible surplus labor value from their employees!

Gotcha. Very good point. Basically "Poor CEOs, the top of the food chain sure is hard..."

Projecting, that's what.

All I read was that the needs and the skill-set to meet them vary with scale. There's no class-ism in that unless you project a lot of other stuff into it.

> Any reasonably smart CEO

Sorry but this is extremely naive and I'm sorry that your naïveté has caused you conclude things which are simply not true.

This is a very strong assumption, and a lie that is peddled by the media (Zuck is the greatest example of this) to create a novel story. I'm not saying it's impossible, it's just a dangerous assumption.

For the record, I've interviewed over 75 software founders (CEO/CTO) and I can tell you right now that there is nowhere near even 50% of them who have the ability to scale their companies.

Furthermore, there is a reason many of my clients (PE investors) replace management teams who cannot scale businesses and when they do, they reap financial awards as a result. So my conclusions are based on following the money...

Of course it’s less than 50%; that doesn’t mean that it’s a negligible proportion, either.

Analogy: (far) less than 50% of military officers have what it takes to be a general/admiral/other “logistics at scale” type. But yet still, military officers with experience commanding in smaller engagements are the only good source of those high-level strategists. If people couldn’t grow into roles requiring more responsibility of them, militaries would be “making” admirals and generals from scratch.

Does that sound feasible? If not, then why would one be able to “make” a Fortune 500 CEO out of anything other than an existing, experienced SMB CEO?

A really good analogy. And it is worth noting that in the UK British Army, junior officer training is nearly a year long course, and then promising Colonels go through a Higher command training course to prepare for Generalship.

Maybe, just maybe, Harvard Business School is the place for teaching better than 50% to do well and expand.

https://en.m.wikipedia.org/wiki/Higher_Command_and_Staff_Cou...

What do you think are they key deficiencies in most founders that prevents them from being able to scale their businesses? I'd be very curious to hear.

(Are they unwilling to delegate and trust others to make decisions? Overly focused on trying to be involved in and influence everything?)

And likewise, I can tell you from experience that nowhere near even 50% of the people brought in that supposedly have the ability to scale a company don't really know anything other than just enough pattern matching to cosplay as Jack Welch in front of investors, and how to minimize their blame when shit hits the fan.

> I can tell you from experience that nowhere near even 50%

I know way more people whose money is on the line that will gladly take that bet.

This sounds like vague hand waving to me.

What direct experience do you have with this to justify that claim?

Obviously sometimes people brought in are nothing more than sacks of potatoes that make noises, but I think its insulting to the intelligence of everyone involved to believe even a significant proportion of people fall into that category.

It's just a different set of skills to the ones for 'technical founder' and involve things like... organising, collaborating, compromising and socializing... you know, grown up things that leaders are expected to be able to do without screwing up.

Scaling a company is lot more than just scaling the tech.

I'm not sure how you jumped from 'similar levels of people brought in also don't have the skills to scale a company and simply optimize for the signals that the people who hire scalers look for' to 'scaling a company isn't a skill and the only things that matters are code.'

I don't believe you unless you can give me a real statistic

Based on what? All of you here saying "they don't have what it takes" have yet to elicit one actual skill that a founder could not _learn_ as the company grows.

There is nothing inherently special about someone who runs a small vs. big company, except for experience. I agree with the poster above: classism at its finest.

> There is nothing inherently special about someone who runs a small vs. big company, except for experience.

I think the question is, given that someone can do one of those things really well, how likely is it that the same person can do the other really well too?

It seems obvious that the kind of traditional business ceo that investors often want to find to replace a founder is not going to be successful as a starter. Let's just accept the common assumption on that one.

I'd argue that what makes a great startup founder often makes a terrible leader for large corporations. I think every great founder has a reality distortion field (https://en.wikipedia.org/wiki/Reality_distortion_field) which projects their vision, keeps them pursuing seemingly bad ideas in the face of insurmountable odds and allows them to overcome the biases and assumptions that dominate so many business decisions.

Some leaders expand and adapt this to build great, crazy companies: Steve Jobs, Elon Musk, Zuckerberg (maybe). Often they also create a cult-like, maybe toxic, certainly messy corporate culture that still somehow succeeds.

Others can't do this: Travis Kalanick, Elizabeth Holmes, Parker Conrad. They let their "startup founder, growth at all costs, thumb your nose at the rules" attitude destroy their company culture and embroil them in ridiculous, easily avoidable legal issues.

In my experience, a corporation's culture is the manifestation of the most extreme personality traits of its leaders. There is a time, when crossing the chasm, that the exact traits that made the company grow rapidly and the founder extremely successful are the ones you want to suppress to create a stable, efficient and effective collaborative endeavor involving thousands of people: a corporation.

I think (and this is due to some reflection on my own ideas) that if you reject the idea that there comes a time when a company should abandon "burning vision" in favor of stability, process, predictability, and broad appeal, you're actually rejecting the idea of a corporation as a hierarchical system of power deployed to efficiently organize the work of thousands of people. And I'd probably agree with that. The traditional corporation needs a traditional ceo but it's a damn shame that every revolutionary startup has to end up as a traditional corporation.

I think what you're describing is the machinations of a capitalist society gone off the rails. Founders are focused, and what you're saying is they're focused _to a fault_. I don't think there's any fault there. Not every company has to continue growing indefinitely. That's insanity. The problem is the shareholders or boards who demand indefinite growth, and do so at the cost of their products, the most devout customers and their culture.

In other words, you can't use a broken system to criticize leadership.

First, it's not about whether they could learn the skill but about whether they will learn it.

Second, if they do learn that skillset as the company grows, then they would be a great man/woman for that job.... at their next company. For this company, you'd want someone who has it right now, not someone who'll learn it halfway, from costly mistakes.

Nobody is suggesting that a large company CEO is "special." It's a different set of skills, that's just a fact. And, the idea of a CEO of a rapidly scaling company to learn through on the job training isn't always a good one. People learn by making mistakes and the marketplace isn't very forgiving -- too many mistakes and the company is toast.

There's nothing classist about it. You could train an army corporal to become a general. But you can't just take an army corporal all the way to the experience needed to become a general in a year or two. It takes generals 20+ years before they have the skills to lead thousands of soldiers. To suggest that any reasonably smart corporal can go from leading a squad to leading a division in a few years of on-the-job training is just ludicrous. If I'm wrong, then please provide some data. Actual data, not anecdotes.

Nobody is saying that a corporal can't ever become a general. The suggestion of classism is implying that a corporal can't become a general because he's not of the right class. Of course a corporal can become a general (at least in the US Army) -- which is the very opposite of classism. We're just saying that a corporal doesn't become a general through a few years of on the job training and that a corporal doesn't become a general while on the battlefield.

What I'm saying is that a good large-company CEO is gets those skills over years of experience. If a founder-CEO slowly scales, then perhaps on-the-job training would work. But when an org goes from 10 people to 1000 people in a few years -- it would take an extraordinarily lucky person to be able to handle that transition without prior experience.

"I can tell you right now that there is nowhere near even 50% of them who have the ability to scale their companies."

You assume that they all want to scale. There is a place for us boring people. We are the ones who do something well and are reliable. We don't blow the profits on fancy cars and houses. We have shit loads of cash in the bank at all times relative to t/o. In 18 years of trading there was one month that me and my partners didn't pay ourselves - to cover employee salaries without creating an overdraft.

Bugger scaling - I'm happy to sleep at night. Some of our customers are amongst those that you will almost certainly have heard of, regardless of which country you live in.

Everyone is too busy believing they are different to hear what you are trying to tell them. Maybe after 20 years in the Silicon Valley meatgrinder they’ll hear you.

Growth for the sake of growth sounds awfully like a bad virus.

> You assume that they all want to scale.

If they've taken outside investment based on a business plan of scaling, then it doesn't matter whether they want to scale. They either need to do it, or hand it over to someone who will.

I love nicely sized companies that are profitable and long term viable without the need to chasing after constant growth, but you only get to do that if the investors agreed to it. That's much easier when the investors are just the founders.

> For the record, I've interviewed over 75 software founders (CEO/CTO) and I can tell you right now that there is nowhere near even 50% of them who have the ability to scale their companies.

What you just wrote does not contradict maerF0x0's assertion. If you can recognize what makes someone capable of scaling a business, and what makes someone incapable, then what prevents you from teaching those skills?

I agree with you.

I work for a Startup in valley, and I agree. Our founder CEO was fired recently. Why, you may ask?

Because he was a really poor Manager, so poor that even Senior Engineers could see it clearly. Luckily board came to senses and replaced him with a proven CEO.

Sad but that is the truth, to scale you need different skillsets and off course being a good Manager is just one of those.

One need only look at Soloman's posting history on this site to see why he was the right person to lead Docker during its inception and the wrong person to lead Docker long term. It was clear in the discussions over Docker that Soloman was passionate about the technology he'd created and had a very particular vision for its future, which was often at odds with the way that posters here saw Docker and wanted it to evolve. But Soloman seemed to lack the ability to be diplomatic about it and showed some poor judgment in what he said publicly. That led to quite a few public spats that reflected poorly on Docker.

It's absolutely true that some people can evolve from a leader of a small company/team/project into a leader for a much larger one. But it's also absolutely true that not everyone can and that the skills to lead that larger endeavor are significantly different from the skills needed to get something off the ground. It's likely that Docker would never have gotten off the ground without someone like Soloman to take a very opinionated stance and drag people towards his vision. But if Docker is to continue to succeed, it now needs to expand its reach learn to be what people need rather than convincing people that they need Docker. Not being on the inside, I can't say for sure whether Soloman is capable of evolving into a leader that can move Docker in this direction, but the superficial view from the outside suggests to me that this was probably the right decision. It's no knock against him as either a leader or a technologist, it just feels like it's not the best fit going forward and his efforts would be better spent on something new that pushes a new vision forward.

I remember when he took the time to hop on my dinky little blog and comment on one of my early Docker posts. I didn't even realize who he was until I saw the name next to the keynote speaker at the first Dockercon. :-)

A lot of research has gone into describing this "classist lie". It is not simply something that the parent stated as a truism. Organizational theory seems to suggest that founders are, as a general rule, not the best CEOs for a maturing company.

See, for example,

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

and its sources.

>Stage 5: Decline : This stage initiates the death of an organization. The decline is identified by the focus on political agenda and authority within an organization,[4] whereby individuals start to become preoccupied with personal objectives, instead of focusing on the objectives of the organization itself. This slowly destroys the functionality and feasibility of the entire organization.[3]

Wow. What if the organization begins in Stage 5?

The question is not whether or not they can learn, but whether or not they did.

History seems to disagree. Growth has killed many companies.

To be really good at something it is necessary (but not sufficient) to be interested in it.

Running a company of 10 people is an experientially different thing than running a company of 1000, just like collaborating is an experientially different thing from doing stuff on your own and playing video games is different from using Excel. Being interested in one is no guarantee of being interested in another.

A classist lie? What are you talking about?

Statistics are pretty clear in the failure rate of founder-CEOs when they scale past certain sizes. Even Zuckerberg -- if it weren't for Sheryl Sandberg, Facebook wouldn't have survived as well as it did. Zuckerberg didn't know the first thing about running a large organization, I might argue that he still doesn't -- but when a company can basically print money, there's a huge margin of error afforded to Zuckerberg-type leaders. If Zuck where competing in the energy business, banking or autos, he'd have been long since buried.

Steve Jobs also wasn't particularly good (managing a large org) during his first stint at Apple -- it took a lot of maturity and learning at Next and Pixar before he was really ready to lead Apple.

Mark Benioff is another founder-CEO that did really well -- however, he was an Oracle Vice President before starting Salesforce. He was both incredibly smart but also had experience within a large organization.

> Any reasonably smart CEO can learn his way from 10 - 10000. Yes, (s)he may not be capable of 10k on day one, but they can grow along the way.

Citation needed.

The list of successful founder-CEOs who have scaled to huge orgs is a very short list.

All of these people had to learn their way up the scale. I see little reason a founder CEO couldn't learn their way up the scale too.

> a founder CEO couldn't learn their way up the scale

OK, but many (most, if others' claims in this thread are to be believed) of them don't do so. That contradicts your original claim, unless you want to say that many/most CEOs are not "reasonably smart", which is dubious at best and armchair quarterbacking at worst.

This is a good point. What's the pragmatic difference between can and does. Others have pointed out a good point too that many CEOs are so busy focusing on other things that they fail to do the learning they need to, effectively falling behind the needs of the company.

I mostly am rejecting the knee jerk reaction of "Company has grown 100x and now requires a new CEO" ... Who knew this would spark so much discussion?

I disagree with the number of employees making things drastically different. 1, 10, or 10,000, it doesn't matter as long as the leader has the fundamental organization skills need to control groups of people. It's easier to hide the lack of organization in smaller groups, but the end state will be the same regardless.

"Management of many is the same as management of few. It is a matter of organization."

With 1 to 10 people you know what exactly who does, his weak and strong points etc. They also know you personally and possibly trust you. You also manage by talking to people personally and you have direct control over pretty much everything you decide to. You see impact of your decisions pretty fast and directly. It is easy to make exceptions and situational decisions.

With 1000-10000 you never seen most employees faces. You have no idea what they do whole day. They don't know you either and likely don't trust you. You manage by making general rules and by talking to few people directly under you. You dont see impact of your decisions directly, you have to guess from hints and long time.

Yes, this is the point though. Managing a company of 10,000 is harder than one of 10.

I agree. I've seen time and time again founders having a certain vision but lack a strategy for getting there. You then get a whole other set of customers that are very different than your early adopters. The priorities in company day to day operations will change as the company morphs into a larger, faster growing business. You then get a whole other set of customers that are very different than your early adopters, and that present unique challenges and different approaches in tackling them.

"There is a reason why most startups don't go big"

I think you are absolutely correct.

I am the MD (CEO elsewhere) of an IT consultancy that is 18 years old. We have around 20 staff and t/o ~£1.5M. I like to sleep at night, I do have ambitions but they'll never be cool. Our business model will never set the world on light but we are one of many companies that you will never hear of that keeps the world turning. We are the boring lot that worry about backups and security. We host some customer systems at our place and run a near comprehensive monitoring system that generally warns us pre failure. We also have a few other skills around the place.

There are other reasons why startups don't go big and that is by design. I do not want my firm to go big. If we hit around 50 staff then I will break it up into bits (budding).

Did I mention that we are boring? I like boring.

Yep! Boring is always best. :)

I like what you said here. Keep a business for 18 year old is a success, and I can only imagine is challenging in many ways, could be boring in few. I have seen greedy consultant companies cutting corners here and there. Caring about backups, security and monitoring is great attitude in my opinion. Keen to hear more story about your venture.

Napoleon Bonaparte (might have) famously described Britain as a "nation of shopkeepers". I spend a lot of my time trying to live up to that ideal! I'll try to keep this brief ...

I was (1997-2000) a contractor in the IS department for the really large aerospace company in Yeovil, Somerset, UK (it's changed hands a few times). I eventually got to the team doing server management, app delivery etc ie third line. IS was due to be "facilities managed" and two of my colleagues approached me to be the MD of this thing they'd cooked up. The idea was we would go external and resell ourselves as a unit back through the on site supplier of IT/IS services. The supplier chosen by said aerospace company used to have a short name that matched the name of a famous sauce from these shores - it's brown (the sauce) and starts with H.

We rented a small unit down the road (proper rural around here) and had an old PC that ran Win 98 as a file server etc etc. We put in a ISDN BRI link (2x 64Kbs-1) for internets. We had one customer.

We moved on gradually and learned. Our first employee stayed for around eight years and then moved on - still in touch. Employees 2 and 3 (poached off a local travel agency - terrible pay there but we could better the pay despite being a little setup: E2 stayed for 10 years and moved on and sadly played silly buggers with customers - nothing too detrimental but still a bad taste left behind. E3 still with us, complications recently but all good now.

By 2013ish we now have around 15 odd employees. We have moved to a bigger pair of units in the same area that we started from. We now have modern systems involving (small) racks, virtualisation etc.

The norm for most companies is to rent but I wanted to own my premises. My thinking was that we could do more stuff. We looked for a while and ended up buying (story short!) an NHS clinic in Yeovil which is the largest town in the area.

We now have five internet connections - a leased line (100M) and four FTTC (ADSL 80/20) and a pair of pfSense HA routers running on Dell R310s (to be ditched soon - they can't do AES-NI) with a lot of IPSEC VPNs and OpenVPNs and 10ish internal VLANS. We have a proper computer room with proper gear. We also have a lot more customers these days. We are now ISO 9001 and 27001 registered and lots more stuff.

Every day I love going to work. Don't get me wrong - it's still hard. I came into work the other day and found an employee in tears. I sent them home (obviously, I can't say too much here). Two of our employees have moved on and then come back again (one of them was once a lover of one of my business partners). Thanks to our past involvement with a large firm whose name is saucy, we have several customers these days that you will probably have heard of.

One thing that I think we have is that we have three Directors - a Triumvirate if you like. Any two can gang up on one - it works very well.

I'm a very lucky person. I'll never be a millionaire but I'm very comfortable as far as I'm concerned. I'm a shopkeeper and fucking proud of it.

Cheers

Jon

CEOs that succeed also have the correct burning vision. CEOs with a flawed vision often drive companies into the ground. I'm not talking about anyone specific here, just noting that many founding CEOs hold too tightly to their own personal vision rather than adjusting it to the market's demands to create a solid direction.

You can't have explosive growth forever. At some point, you look around and realize that you don't need to worry about coming up with big ideas anymore and that what you've developed is mature enough to be a real revenue generator. Hykes clearly isn't the kind of person that likes to sit back so he's off to find a replacement, the guy that will be really good at streamlining what Docker does (and ensuring that the stock does well).

> Companies that succeed are driven by burning vision, not by day-to-day operation people.

What is the evidence for this? We can throw anecdotes like Apple on one side and Unilever on the other, sure, but is there any study saying e.g. "Of these 100 promising companies founded in 1990, x% of the founder-led ones were successful by 2010 whereas y% of the outside-CEO-led ones were?"

Making a bold claim like "Companies that succeed have X cool-sounding behavior" is the realm of business books in airport bookstores. It would be nice to take a more data-driven approach.

Also, the vast majority of startups do not lack for drive & vision, but most will fail. Vision may be necessary for a new company, but it certainly isn't sufficient.

> What is the evidence for this?

GP's burning vision, of course.

I once had my vision burn so hard my landlord charged me for the floor paneling.

Noam Wasserman (Harvard, USC) studies exactly this in a paper called "Rich vs King". As you can guess from the name, the conclusion is that bringing in outside generally management maximizes the expected value of equity.

https://hbr.org/2008/02/the-founders-dilemma

Thanks!

One thing the article brings up is that "success" may not be synonymous with "financial success". If you have a specific burning vision for a thing that should exist in the world, it might not be the most profitable thing, and the founder staying on is probably a good way for exactly that product to exist - but an outside CEO might choose to compromise the product vision and significantly expand the potential market in doing so.

Very true!

I suspect involuntary founder replacement is much more common in venture-funded companies. While your, my, and the founder's definitions of success may not be synonymous with "financial success", many VCs and their LPs may not feel the same. Obviously it's tricky to fully generalize, but the point is that VC can be expensive if founder and investor motivations aren't closely aligned.

It’s the standard HN cult of the founder post. Goes right along with the idea that early employees deserve pretty much nothing because “it’s the founders taking all the risk”.

Agreed. It's amazing how many people (including many here) prioritize being CEO of a successful company over being the founder of a successful company. Attaching your Self worth to the continued adoration of others in the business community is silly and dangerous. I know people who founded hugely successful businesses and their moving on from being CEO played a big role.

It's also entirely possible that Jobs would have run Apple into the ground in the mid 80s. Apple III and Lisa were expensive, unmitigated failures, and the Mac arrived as Jobs was leaving. Would Jobs have gone along with the high-price/high-margin strategy of Sculley/Gassee?

Was about to say exactly that. Yes, Jobs 2.0 was a massive success at Apple, but it's not clear to me that Apple would have survived the 1.0 -> 2.0 learning curve.

Companies that succeed are driven by burning vision, not by day-to-day operation people.

They are not mutually exclusive; thinking that they are is a recipe for failure.

Both are critical. You have a goal and then you make progress. Neither is enough on its own.

Lets step away from the "company" aspect of it, because docker is also a project.

If this post were by someone named Linus, or Guido, (or Jobs) or any number of other names I can think of - people would be freaking out.

Yet the reaction so far is one thatI can only describe as "meh" - there wasn't a scandal (uber) there isn't a failure (intel) and people mostly (with exception) like the product.

I wonder what the future holds for docker after this change.

Also that art at the end, looks great but all I could think of was "fail whale" when looking at it (and not docker).

The people that have both are outliers, but massively successful: Bezos, Jobs, etc. But that just shows the importance of having both types of leadership in a company, because the type of drive that grows an innovative company isn't the same as what keeps them in the top spot. This isn't a simple binary variable, just about every company has had or will have to find their balance on this.

It may not be glamorous, but no matter how innovative or successful, someone (many people really) have to have an eye on day to day operations. A CEO should have vision, yes, but their job isn't just to have vision, it's to share that vision with the people under them, the day to day people that have to be in place, and will make the best day-to-day decisions only if they understand and share that vision.

Disagree. Counter example: Red Hat (disclaimer: I’m at Red Hat since almost 13 years).

Founded by realistic visionaries (Bob Young and Mark Ewing). Bob was smart enough to bring in Matthew Szulik, who gave Red Hat structure and growth. And Matthew was smart enough to hand over to Jim Whitehurst.

Red Hat is nowadays a smooth running company with 64 quarters of growth, with a portfolio that goes far beyond Linux.

Execution is key. Having passionate employees a must.

This is a good point. I think that a company needs burning passion at the team level in order to succeed, and calm reasoning heads handling the day to day to allow those passionate people to continue to do what they do.

Bob Young founded Lulu.com, the ebooks and print books company, before Red Hat, IIRC. That was an early pioneering ebook and self-publishing company startup, I think.

Lulu.com was founded in 2002 after Bob Young left Red Hat.

Okay, I got that the other way around, thanks. Seems there was quite a gap between the two foundings, too.

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

https://en.wikipedia.org/wiki/Lulu.com

Looked up Bob Young on Google:

https://en.wikipedia.org/wiki/Bob_Young_(businessman)

and it seems he also was a co-founder of Linux Journal! Didn't know. Cool. His work has had quite an impact.

I vehemently disagree, burning vision is wasted when you don't focus on the day to day.

The day to day is what actually leads to revenue and hopefully actual profits. You can only burn vc money for so long until you actually have to show some sort of return.

Counterpoint: Google brought in an established CEO to get the nuts and bolts of being a big company working correctly. I would say that Google is pretty successful.

I think that also points out the value of engaged investors and board. I don't know the details but I am under the impression they really facilitated that process -- recommended the idea, found candidates including Eric Schmidt, did the whole thing without disempowering Larry & Sergei, etc.

Because the Google founders were PhDs with more technical interest than business.

Plenty of companies are founded by people who aspire to lead large companies.

At least for fundamental things like containers, Docker's current founder definitely is not a match for many other people in technical capability and experiences.

The later part is more severe.

It called facing reality and realizing that certain phases in a company's growth require having someone with real world experience at the helm.

It's the reason why John Doerr brought in Eric Schmidt as Google's CEO for adult supervision. Good investors, in addition to bringing cash, also bring with them accumulated experience from previous rodeos, which means they'll be able to recommend a few qualified hires for the up hill battle ahead, if it is clear the founder is lacking in requisite experience.

Ideas get you to the 50 yard line, execution gets you into the end zone.

More like the 21 yard line?

I disagree. This is EXACTLY why roles such as "Chief Operating Officer" and "Director of Operations" exist.

You don't make your company succeed by giving up control. You hire people you do control who have the skills you lack.

And those skills may include taking the company in directions that you may not agree with. What then?

You have to be smart enough, and humble enough to recognize when someone else's vision is what your company needs at the moment. I'd imagine that a good CEO would be able to both simultaneously provide the direction of a company while also accepting input, on how to actually get there. Navigation, and steering are two different things

What if the skill the founder lacks is in leadership or vision? People like Bill Gates and Mark Zuckerberg who can code AND lead are the absolute minority. I know more than a few founders who, while charismatic, absolutely do not have the temperament or skills to grow the business beyond 50 or people.

>> you dont make your company succeed by giving up control.

Of course you do. Every time a new hire is made the founder gives up some control. The secretary now chooses the paper for the copy machine. The sales guy now enunciates the detail of the offering to the prospect. The engineer codes in her specific way. Growing a company is ALL about giving up control in exchange for accelerated growth of the business.

That's exactly what is being done - the founder stays on the board as a shareholder and controls the decisions on overall direction, goals and the vision of the company, and hires an executive that has the skills required for management of a large company and will execute these goals.

If at some point, you end up being CEO but have delegated all the real work, you are a waste of space. There are roles like Chairman for people who aren't essential to day-to-day leadership.

"Companies that succeed are driven by burning vision"

lol.

/not even close.

Not sure why this was downvoted. It's a good (and correct) point.

It doesn't really add much to the conversation, and it comes off as rude.

I disagree. At a certain point, things get very serious. When companies become successful and deal in billings in the 100's millions or billions, it's time to get a more experienced team.

It's the ethical thing to do; it's a fiduciary duty of leadership to do so.

He may be stepping aside because what's required to lead an enterprise company is not something he's capable of or interested in doing; it's a radically different thing then created a company from nothing.

I think Shykes stepping aside is an extremely professional move and a hallmark of great leadership. Honestly, it's not something you see often in tech; many founders hold on for dear life and companies suffer greatly because of it.

Agreed. And leading and inspiring a team is done by having a burning vision, communicating it to the team, getting them excited and making it their vision too and doing that on a daily basis.

Day to day operations can be handled by COO, CFO, CMO, VP and other lieutenant roles. Why not hire Steve Singh as chief operating officer? Why not even lower down the totem pole? Is it because there's a fear that 'talent' of that quality may not take the job without a CEO title? Money can solve that. If there's an ego in the way, perhaps that's a problem.

I've scaled our business from 2 people to 10 to 20 and now to 30 with high 8 figures in revenue. I'm quite comfortable that I can scale it to any number. The reason is that I am very comfortable hiring people that are smarter than me and putting them in leadership roles in the organization, stepping back and trusting them and heavily relying on them to do an excellent job.

Stepping back and trusting takes an incredible leap of faith as a founder. You also need to be comfortable asking Fortune 5000 execs to work for you in a VP or C suite role. You also need to be able to pay them. If you can do those three things, you can assemble a team of lieutenants that will run their business divisions, provide you excellent visibility in all areas without over communicating and you will be able to scale into infinity as they replicate your strategy for scaling by hiring trusted lieutenants.

It also helps if you know how little you actually know, can be completely honest about that with your exec team and ask them for help and input and trust them to give you what you need.

I haven't experienced this, but it seems to me that firing founder CEOs and 'hiring a mature CEO' is conventional VC wisdom and it's based on a pattern they've seen work. e.g. Cisco. Once you raise big VC, I'm not sure that as a young founder you actually have a say unless you specifically build controls into your funding agreements. In FB, Zuck controls about 60% of the voting rights. He's 33. You have to ask yourself if he'd be CEO if the voting structure was different.

It seems that a lot of people on HN have a visceral negative reaction to founders stepping aside in favor of bringing professional management. Ousters aside, I wonder if it has to do with value creation versus value capture phases of the company, with people here being generally more interested in creating value than capturing it?

Kinda funny that the page is down ... you'd think for a company that helps everyone scale their services this would not be a problem :-)

I'm not sure resource scaling is what's needed here. Just good ol' caching. Shameless plug: I built Cachoid[0] for caching as a service (leveraging Varnish mainly) to help with situations like this.

[0] https://www.cachoid.com/

Maybe state that it is a Varnish configurator. I initially thought it was a new caching proxy, thinking "Why would anyone need that when they have Varnish?".

Also, adding caches is still resource scaling.

Edited to mention Varnish.

Also, adding caches is still resource scaling.

This is a semantically correct. Caching can indeed cut down on resource usage. It isn't always a scale-up operation.

My point is that a caching server is a resource like any other. Adding additional backend capacity, and adding additional caching capacity are both "adding resources"/"scaling up".

Doing smart things to do more with the same resources, do the same with fewer resources, or by replacing expensive resources with cheaper resources, is optimization, not scaling.

We're being really pedantic right now, though. :)

I understand that this is a shameless plug but I've met teams where the first response to any performance problem is caching

If you're reading this and have performance issues look at caches as a last resort, first figure out why things are slow

I'm assuming a traditional stack here but use mysql jet profiler, or SQL equivalent first, then your languages profiler xhprof?

If your queries are slow or impossible make sure your tables are normalised and indexed, make sure opcache is on and xdebug off

Mysql out of the box is very conservative with its resources, the limits can often be raised and the web server can often be configured to be faster, make sure you're sending the correct http cache headers

There's so many things to consider with performance Please don't denormalise your data slap it in redis/ use varnish and call it a day

No cache is free, eventually you have to pay the pied piper when you do make sure you can afford to have never used the cache at all

Some of my hardest bugs have come from out of sync caches & trying to build performant data architecture around systems that had data stuffed into a redis black hole aged for years..

Also if you're doing key scans in redis, you're doing it wrong, it's possible to work around with indexes you have to manually maintain, but it's not worth it

Lots of good common sense advice and practical experience there. I've had my fair share as a devops/sysadmin working on crazy systems (more worse than better haha). So it's unsurprising how common sense isn't always put into practice.

I do believe caching to be the proverbial Swiss Army Knife of performance. Some of its tools can be used properly and others as band-aid / stop gap. Even band-aids serve a useful purpose. But I can definitely see how a system can accrue technical debt when over leveraging caching.

But to believe in prevalent, perfectly built systems is to deny the status quo of the landscape. When you democratize development like WordPress has done with its plugins and Node has done with npm, as popular examples, you definitely have it coming.

I dunno, I'd say solve your customer pain first, then mess around with your profiler. If throwing up a cache solves your immediate problem, by all means throw up a cache.

Just make sure it's your first step, not your last step.

I totally agree. Scaling with docker containers is not a silver bullet for a LOT of problems. Optimizing your boring/old tech stack can much more often a better investment than betting or (even worse) switching on docker as your foundation.

The author Solomon Hykes, user shykes, announced DotCloud's Docker was going opensource here on HN almost exactly 5 years ago on March 26th, 2013.

https://news.ycombinator.com/item?id=5445387

Aside for the near coincidence, I find reading historical first release posts very enlightening!

I'd be happy if any of you would share first-reaction posts and chatter and more on Solomon's marketing-first approach on making Docker a tremendously successful opensource project.

Like a lot of tech folks, I've only interacted with Solomon on the internet, but in probably every interaction, I've come away from it feeling smarter, or more capable, going all the way back to the very early days of Dotcloud.

I understand the realities of working in a corporation, and the demands on C-level executives, but it's hard to imagine any company that is improved by his departure from it.

Fare well, Solomon. Eager to see what you've got planned next.

Do you work for docker? If not, there's no need for farewells. It's not my impression that Solomon is leaving the planet :)

Very excited to see how this new C-group executes.

Probably one of the best departure messages I've ever seen.(especially with that image)

Four months ago on HN, an anonymous account posted "SHykes has left Docker" [0].

[0] https://news.ycombinator.com/item?id=15650335

Why’d they do that?

Another au revoir post [0] by Jérôme Petazzoni (also mentioned in shykes's post) a little over a month back.

[0] https://jpetazzo.github.io/2018/02/17/seven-years-at-docker/

I remember that. I was one of the users asking for the source. I wonder who the OP of that post was.

I was able to get the page to load, here's the text of thet post: "Today I’m announcing my departure from Docker, the company I helped create ten years ago and have been building ever since. A founder’s departure is usually seen as a dramatic event. Sadly, I must report that reality is far less exciting in this case. I’ve had many roles at Docker over the years, and today I have a new, final one – as an active board member, a major shareholder and, I expect, a high maintenance Docker user. But I will no longer be part of day-to-day operations. Instead, after obsessing for so many years over my own ideas, I am rediscovering the joys of putting myself at the service of others – my friends, my family, and the brilliant entrepreneurs I’ve been lucky enough to advise and invest in over the years. Over the coming months I plan to use my experience to help them in any way I can.

This transition is simply another chapter in a long story of change, growth, hard work… and a lot of luck.

Ten years ago, I quit my job, returned to live with my mother in Paris and, together with my friends Kamel Founadi and Sebastien Pahl, started a company called Dotcloud. Our goal was to harness an obscure technology called containers, and use it to create what we called “tools of mass innovation”: programming tools which anyone could use. I was 24 and had no idea what I was doing. We needed a CEO, so that became my new role.

Five years ago, Dotcloud reinvented itself as Docker, around a battle-hardened core of five people: Eric Bardin, Sam Alba, Jerome Petazzoni, Julien Barbier and myself. Soon growth was off the charts, and we hired an experienced CEO to help us sustain it. I was 29 and eager to do my part. Docker needed a CTO, so that became my new role.

Today, as I turn 34, Docker has quietly transformed into an enterprise business with explosive revenue growth and a developer community in the millions, under the leadership of our CEO, the legendary Steve Singh. Our strategy is simple: every large enterprise in the world is preparing to migrate their applications and infrastructure to the cloud, en masse. They need a solution to do so reliably and securely, without expensive code or process changes, and without locking themselves to a single operating system or cloud. Today the only solution meeting these requirements is Docker Enterprise Edition. This puts Docker at the center of a massive growth opportunity. To take advantage of this opportunity, we need a CTO by Steve’s side with decades of experience shipping and supporting software for the largest corporations in the world. So I now have a new role: to help find that ideal CTO, provide the occasional bit of advice, and get out of the team’s way as they continue to build a juggernaut of a business. As a shareholder, I couldn’t be happier to accept this role.

As a founder, of course, I have mixed emotions. When you create a company, your job is to make sure it can one day succeed without you. Then eventually that one day comes and the celebration can be bittersweet.

It’s never easy for a founder to part ways with their life’s work. But I realize how incredibly lucky I am to have this problem. Most ideas never materialize. Most software goes unused. Most businesses fail in their first year. Yet here we are, one of the largest open-source communities ever assembled, collectively building software that will run on millions of computers around the world. To know that your work was meaningful, and that a vibrant community of people will continue building upon it…. can any founder ask for anything more?

I want to thank from the bottom of my heart every member of the Docker team and community, past and present, for making Docker what it is today. Thanks to you, this founder’s bittersweet moment is mostly sweet. We have built something great together. I look forward to seeing where you will take it next.

Happy hacking,

Solomon"

Having read this I wonder what really went down. I get that his shares are fully vested, but personally I would not walk away from such a role at a company with this kind of growth and mind share.

Nothing terribly interesting needs to have gone down. Someone who is the right fit to lead a company from a scrappy three-person startup to a large established tech company, and enjoys it the entire time, is a unicorn among unicorns.

Isn't it really hard to know what you would or wouldn't do in such a situation? What if he's just kinda tired, you know?

He’s in a position where the rational and realistic response to being tired could be as simple as delegating more and saying “from now on I’ll be working 3 days a week”. Our CTO doesn’t even show up for weeks at a time.

That's probably not the best solution for both the person (you hardly get any mental rest just by working a bit less) and the company (which probably needs a CTO, otherwise there wouldn't be one).

I mean if you have "enough" money and are not enjoying the job, I could see stepping down.

He’s put his heart and soul into the company for 10 years, is wildly successful, and doesn’t think he’s the best person for this next phase.

I don’t think any back alley drama needs to have happened.

Maybe he doesn't want to be that person?

Founding a company is much different from running it.

I'd consider founding exiting for the first years, but later rather boring.

More than that, he's put his energy in the same company for most of his professional life.

Nobody wants to be a one trick pony. Even if you are successful beyond your wildest dreams, people want more than one act in their lives.

If I had founded docker, I'd be retiring now[1], because I've done other things and this might feel like the cherry on top. But he's only 34.

[1] If I knew how, which I don't. At best I'd switch fields and start over on something else I care about.

Why? He has the money and the product does what he wants. If the community he's help build around the product is in good shape, why do anything else?

If everyone had that kind of mindset, Elon Musk would still be running PayPal.

Not a good example. Musk was ousted from his CEO position there.

And many of us would. I'm not a founder or a visionary, but I often commented of my former CEO "If I had hundreds of millions of dollars and was in my mid-70s, I sure as hell wouldn't still be running a company, I'd have retired ages ago" ..... followed closely by the comment "and maybe that's why I don't have hundreds of millions of dollars!"

Not everyone has the desire to persist in that same role, particularly as the company evolves with size. For some, the money allows you to walk away and pursue passions that are independent of a need to make money. And/or maybe his passion is starting a new project that MIGHT be a business that MIGHT make money, just like Docker once was.

Docker is a reflection of some of the side effects of HN. Self driving cars and AI are other tech being pushed in a misleading way. All these aggressively promote SV funded companies without proper technical scrutiny.

The lack of scrutiny means questionable technical decisions and complicated tool chains that add layers of complexity with little benefit are passed on to users with little knowledge of the trade-offs involved.

In a way the original promise of containers of being simple and portable have been completely lost to complicated wrappers like Docker and Kubernetes and words like devops that promise a lot but deliver an out of control web of complexity.

Google and Facebook have an army of networking, storage (they even hired the btrfs developer), security and scalability experts. It's not 2 engineers doing devops.

This also sets perverse incentives. Wrappers like Docker and Kubernetes monopolize all the attention and funding while the core open source projects that they are critically dependent on like LXC, Aufs, Overlayfs, Btrfs, Dnsmasq, Nginx, Keepalived, Consul, Quagga, Bird, open source networking tech, container tech in the kernel, apps are all hidden from view.

Docker initially wrapped the LXC project without proper credit and there was a lot of misinformation about what LXC is, with zero pushback here. Many commentators even today are not clear about containers and think Docker invented containers. And this after thousands of posts about containers here.

> Wrappers like Docker and Kubernetes ...

How's Kubernetes a wrapper? Wrapper on what? The ones you listed are dependencies. What projects don't have dependencies?

> Docker initially wrapped the LXC project without proper credit ...

Docker hasn't been a wrapper for lxc for a long time (at least 3+ years, which is a long time in tech world). Also, the debian package at the time was called lxc-docker.

> Many commentators even today are not clear about containers and think Docker invented containers.

Misguided people are always going to be the majority, especially in the tech world. People actually working with these technologies, they will bother to figure out the differences.

Thanks Solomon for creating the inflection point of the container revolution.

I'm a bit puzzled about where docker as a company has to go. But the enthusiasm that was unleashed is without par.

I'm going to second this - docker has literally opened up pathways that very well be underpinning 80% of all new projects.

Because of Docker I can bootstrap new employee's working environment in 5 minutes - everything is a container except they thing they are changing, commit, see.

We still use Vagrant - because it's cross platform. And run docker inside that.

Back in 2011, Solomon was nice enough to get lunch with me along with a few other DotClouders to talk me through some scaling issues my startup was stressing over. They were all super knowledgeable and helpful. We were able to get a handle on things thanks to their advice. Thanks so much and best of luck!

I'm only interested in 2 things. a) Is he happy, was the journey worth it? b) How much money did he make?

So long as he is happy and he didn't expose himself to any financial ruin. It's all good.

Very well written. It has to be tough to walk away from something with as much trajectory. It also takes someone very self-aware to know when they are helping and where their talents can be better utilized.

Solomon being French - and considering the title - my brain automatically played tricks on me and as I read his post's title I pictured him saying "Au Revoir" in its most famous form in France: https://www.youtube.com/watch?v=0obJfkU9xB4 . :D

(It's the official good bye of former French president Valery Giscard d'Estaing as he left office, and it's been used over and over again as a comic element.)

My hat tip to shykes. We studied in the same engineering school and he was part of the teaching assistants' panel doing the interviews for the new batch of candidates I was in. When I met him that day, he was in charge of testing my conversational English level and this interview (amongst others) convinced me to go for this school. He was also one of my tutors/teachers in the first few months, and I eventually became part of the same teaching team. Great guy all around, talented technician and a great educator.

I think he entirely deserves the praise for creating dotCloud and Docker, no matter what people may say about the underlying technology and the lackluster Docker activity lately.

He's turned it into a great product and company, with impacts and snow-ball effects all over the IT landscape.

Back in 2011, Solomon offered his team to help us out with some background job pool issue--mostly Paul. Thanks, and hope you planted enough little seeds along the way that you're able to get back to doing what you do best--making seedlings grow.

I hope it's not the case of enterprise bullying. I still believe Docker has the potential to grow into a much larger enterprise company if only it can navigate through the treacherous waters.

I'm getting a 504 Gateway timeout. Is the HN hug of death really affecting such a huge site?

blog.docker.com runs Wordpress, and the domain points directly at an ELB. Unless they are running some caching layer (behind that ELB but in front of their wordpress processes), they may have to scale out a lot in order to handle traffic spikes, and they may be limited in their ability to do that by DB capacity.

Interesting. Thank you for the clarification!

"Our goal was to harness an obscure technology called containers"

Was that really the goal? My impression was that dotcloud was just a hosting company

yes

Another source: http://www.eweek.com/enterprise-apps/docker-founder-solomon-...

cached link: https://webcache.googleusercontent.com/search?q=cache:vBSgkE...

What Solomon is doing, walking away from the company he helped create, is probably one of the hardest decisions an entrepreneur may need to take. Good luck with everything!

I've seen this before firsthand and it always looks painful. It's easy to look at something from afar and think, "why not just let it go?" But seeing the pain of letting go time and time again makes it seem far more difficult....

docker ps -a SOLOMON HYKES Exited(0) 1 day ago

Sad to see him go, so so grateful for his work!

Unintentionally hilarious that the Docker blog of all companies should have received the ol’ reddit hug of death...

Honest question:

How common is it for a company to flourish after the original sole founder leaves?

Having seen founders forced out by the board first-hand multiple times, I've yet to see things really take off after the departure of such a key player.

I'm curious if, on average, it's really the right move.

It's too bad but this is unfortunately how things work. You may start the company and as you get more and more people on the board that are not on your team your voice gets quieter.

Until the people on the board vote to replace you and the silence is deafening.

Brutal what business has come to

This guy led Docker's marketing-first approach here on HN: https://news.ycombinator.com/user?id=shykes

https://hn.algolia.com/?query=author:shykes%20docker&sort=by...

I put CloudFlare in a similar boat, but CloudFlare seems more tech-savvy (apart from one giant screw-up that cemented marketing-first for me).

I'll always be sad that marketing wins.

What does that mean, "marketing-first approach?" Is this a common distinction in the Hacker News culture?

If I had to pick a more common phrase to research on HN, it would be "growth hacking". By marketing-first I mean growth hacking/PR is priority #1.

I'm not saying anything against this person professionally or personally, it is more a textbook example of marketing hustle as key to financial success. I prioritize technology ahead of marketing, and am currently struggling to flip this around.

PS. I didn't know Docker was YCS10, so props to YC for connecting with this person early. Growth hacking definitely fits into the YC culture.

> I prioritize technology ahead of marketing, and am currently struggling to flip this around.

May I ask why you feel strongly about a technology-first approach?

I can think of someone that tried your exact approach, but it didn't end well. He happens to be a member of this community [0] and he wrote a great blog post [1] introspecting as to the reasons why technology-first approaches don't work in the world we live in.

If you've not read it, I encourage you to do so.

[0] https://news.ycombinator.com/user?id=coffeemug

[1] http://www.defmacro.org/2017/01/18/why-rethinkdb-failed.html

I just love tech a lot more than marketing, probably primarily because I'm much better at tech!

I'm glad that RethinkDB seems to be headed toward a happy ineding as a community, and thanks for the pointer to the write up.

IMO Docker and CloudFlare have lived up to their hype more than most startups, so it seems pretty cynical to complain about their marketing.

Wow. These kinds of communications typically seem trite. But this one genuinely resonated with me. I'm glad things worked out for him. And thanks for Docker!

Am I the only one reading a K8S concession into every Docker headline lately?

I made a big bet on Docker a few years ago, so I know personally that's been the case.

Does this mean that Docker as a company is stable enough to function without one of its co-founders? Or does that mean that VC is taking it over?

It looks like their blog crashed under the load... Here's the full text below.

Today I’m announcing my departure from Docker, the company I helped create ten years ago and have been building ever since. A founder’s departure is usually seen as a dramatic event. Sadly, I must report that reality is far less exciting in this case. I’ve had many roles at Docker over the years, and today I have a new, final one – as an active board member, a major shareholder and, I expect, a high maintenance Docker user. But I will no longer be part of day-to-day operations. Instead, after obsessing for so many years over my own ideas, I am rediscovering the joys of putting myself at the service of others – my friends, my family, and the brilliant entrepreneurs I’ve been lucky enough to advise and invest in over the years. Over the coming months I plan to use my experience to help them in any way I can.

This transition is simply another chapter in a long story of change, growth, hard work… and a lot of luck.

Happy hacking,

Solomon

twitter mirror: https://twitter.com/solomonstre/status/979034154384834560

Did anyone else well up at that final image? It's been a long week...

It's a nice image, but I'm trying to figure out what kind of rig that sailboat has.

I think it can only sail backwards...

Hug of death?

Twitter post here: https://twitter.com/Docker/status/979027618866520064

Title should change. Sounds like Docker is shutting down permanently since the site doesn't load and the title is French for Good Bye.

Haha, I clicked on it in a panick after remembering what au revoir meant.

Yeah same was about to shit my pants...

I dunno, I knew what the content of the post by seeing the title and the domain on its own. People (and companies) don't say good bye in the first person when they die, since they're, uh, dead. They just go "there is ... another ... sky... walker..." and disappear.

Well, 504. No cloud auto scaling?

504... rip

If you read between the lines, there's no mention of Ben Golub (Docker's CEO until few months ago). I am wondering if that's intentional, or... intentional. Or intentional?

Oh, the irony of the docker blog timing out...

Looks like they are using Docker Swarm to host their website.

504 gateway error?

edit Solomon Hykes is departing - https://twitter.com/Docker/status/979027618866520064

http://webcache.googleusercontent.com/search?q=cache:0VT_m5S...

Is it just a conscious joke at this point to put "incredible journey" in any of these types of announcements?

'Incredible journey' is usually reserved for software companies selling out or shutting down, neither of which is happening in this case.

The parent likely picked it up from the Twitter announcement:

"The entire #Docker team would like to thank @solomonstre for creating the containerization movement and his 10 years of service and innovation. Thank you for bringing us all along on this incredible journey – we look forward to continued collaboration https://dockr.ly/2ut6OyO"

https://twitter.com/Docker/status/979027618866520064

Maybe they genuinely thought the journey had been incredible and wanted to express their gratitude to shykes, who after all, built the company they work for?

It isn't a bad or forbidden phrase in and of itself, it just usually sounds rather hollow when users are getting screwed by a service shutting down. But Docker isn't shutting down.

Also, consider when railing against cliched thinking that the 'Our Incredible Journey' Tumblr is by now a massive HN cliche. It no longer sounds insightful or smart.

http://svdictionary.com/words/incredible-journey

Website is 502'ing, here is a the WaybackMachine-Version: https://web.archive.org/web/20180328164640/https://blog.dock...

Site appears to be down but I'm guessing the post is related to this tweet? https://twitter.com/Docker/status/979027618866520064

Very short blog post from Solomon with a little bit of his history at Docker, noting that the company is at the point where he can leave it to run by itself without him.

A founder has retired, nothing as dramatic as the title might imply.

Finally loaded for me. Solomon Hykes is leaving the company.

HackerNews often snuffs out the websites it links to -- the "hug of death". Damn it. Guess I'll come back in 20 minutes.

https://twitter.com/Docker/status/979059103128436736 twitter has a screenshot

I think their blog being hugged to death is a good example of how this technology (read: web backend tech in general, not containers) is still too complicated for people in general to use it for simple cases, like scaling the load of otherwise static content. (It's wordpress, but still) We need solutions that require less resources and smaller learning curves.

I don't know how their blog is hosted but wordpress in general is very chatty with the database backend, and I really don't think the site getting crushed by HN is indicative of fundamental issues with container technology. If it is then there is a long list of technologies that HN has exposed fundamental issues with by crushing their content sites :).

Oh I'm not pointing at container technology. I'm talking web backend technology in general. I mean this is 2018. No well known company's blog on AWS should ever go down at all, unless the AWS service in the region is down. The fact that it's down is not a Docker failure or a Wordpress failure or a Database failure, as there are widely known methods to scale all these things. The problem is that putting the puzzle together is just complicated enough that it's easy to get wrong. If even Docker's blog can't scale, we have bigger issues as an industry.

Fair enough. I took "this technology" to be a reference to Docker given the context.

I'm sorry, is your argument based on the current year? I presume you mean otherwise, but that's how it's written.

Yes, and it's a perfectly fair argument.

Content hosting sites are supposed to host content. That's what they do. If a site doesn't work under unusual but not outrageous conditions, it's fundamentally broken and not fit for purpose.

The industry should be much less tolerant of excuses for tools and systems that perform poorly under load. It's had a couple of decades to get this right now. How is it acceptable that this is even a problem?

Static site generators are an infinitely better choice for making actual blogs. Wordpress has evolved into a CMS similar to Drupal or Joomla with all the available plugins. You can make it into a store or a wiki or basically anything, but you probably shouldn’t. Wordpress was really cool back in the PHP5 days, but Moveable Type and other “old school” static CMS generators always scaled better. Today with node and the dozens of good SSGs, there’s really no excuse to use WP for a blog. The content doesn’t change between page loads, so why would you server-side generate the page dynamically on each page load? The way that it works is fundamentally wrong, and while it’s not very hard to fix with Varnish or WPSuperCache...why not start with the right tool for the job?

Try getting 5+ marketing people to learn git and GitHub sometime and let me know how it goes.

For just the blogging portion, you can just add the blog posts directly through github (the markdown files for posts).

https://help.github.com/articles/creating-new-files/

But yeah, it’s less user friendly. I’m sure someone will make a good solution for this soon.

There are some web services that will run a static site system for you like https://www.netlifycms.org/ or https://www.datocms.com/ .

Gatsby + Netlify + Netlify CMS is super.

Git is not for deployment. Also, hardcoding blog posts in the repo is a dumb solution, and not necessary for a static site. The user built markup should be separate from the code that generates static pages from it.

This has also been my experience when pleading with clients about not using Wordpress... even when they pay me to research SSGs, they seem to always revert back to WP because that's what the non-engineers know how to use.

There are still plenty of CMS options that are built around caching in a way that you effectively get the best of both worlds. I'm a big fan of Grav, which is a flat-file CMS with an optional admin panel.

The aforementioned Movable Type has a pretty graphical interface just like WP. It's just that in the background it generates static files.

It's depressing that 17 years later it still seems to lack a decent competitor.

It just works.

99% of the time for 99% of the people who want it to do something (something realistic), it does it. In order to get people to switch to something like a static site backend it has to be much better than wordpress. It can't be as good, or a little bit better it's got to be waaaay better or it's not worth the time and effort to switch away from something you're familiar with.

If Docker's blog was their core product, I would probably be more likely to agree. The truth is, they don't get ROI from engineering their blog to anticipate or quickly respond to all the traffic of a major announcement, because those are few and far between. The level of complexity is moot because the benefits will only outweigh a very small cost.

They build container tech. That doesn't mean they will benefit from engineering the best solution for everything they do.

I'm not arguing with that, I'm saying it's dumb that we still have to work hard to get a blog to scale.

There is a slight irony in them being an infrastructure company.

WordPress works very well with high loads if you use proper caching (via WP Super Cache, for instance) so the database doesn’t get hit for every page hit.

You don’t need containers or in fact anything fancy for that.

With WordPress in particular containers make some maintenance and scalability issues more difficult to deal with.

Or it's completely unrelated.

Maybe he originally set up the blog using his work login and HR auto-deleted his credentials or account. Sounds like something that would happen around here...

I don't think Docker is too complicated; instead, I think a lot of companies misunderstand the benefit of containers aside from "my application boots faster"

We need more static page generation like Jekyll. It doesn't make sense that an entire blog needs to be dynamically generated all the time. Dynamic contents can be fetched using ajax, but the rest should be static HTML.

Static content can and should be cached, not necessarily pre generated. Thus you can use your site as is and handle caching with whatever you need. It also takes care of semi static (slowly moving).

I've been mumbling under my breath about this since the 90s. The basic issue as I see it is that nobody wants to do the up front work of separating out static, client side dynamic, and server side dynamic. If that work were done then nominal load would be much less to begin with, and scaling up would be much more straightforward.

Maybe I'm misunderstanding your categories, but why would a server care about the distinction between "static" and "client side dynamic"?

A server would not care. It's the devs who need to separate these things.

Mumbling since 90s? Well, if you stopped mumbling and showed the world how it ought to be done, perhaps we would be doing it your way. Show us how.

I don't understand why companies take on hosting their blog and marketing page (usually the homepage) when it isn't their core competency and a distraction for the engineering team to maintain.

It looks like this is true for Docker which seems to be hosting it themselves instead of using Wordpress.com

» nslookup blog.docker.com Server: 192.168.86.1 Address: 192.168.86.1#53

Non-authoritative answer: blog.docker.com canonical name = blog-1556129762.us-east-1.elb.amazonaws.com.

Or use WpEngine or FlyWheel (https://getflywheel.com).

well if a company which tries to sell solutions for high scale , high performance solutions and fast rollout and scaling. hosts their own blog (instead of using a hosted solution) AND can't scale it to their needs, than I would say their product failed. Either because they couldn't use it in this case (because too hard, etc, whatever) OR because it basically didn't worked.

They are a container company, not a scale your blog + scale your database + scale your cache + scale your security company.

Than they should‘ve hosted it somwhere else instead of on their own tech

See my first comment.

When did that become so hard -_-

Because a blog at scale is still a complicated piece of software? You need to implement caching, DoS protection, security - something a hosted blog vendor might have experience in.

What do you mean still? It's about the easiest thing you can do... It was easy a decade ago and I reckon it's easier now. For Wordpress it can be literally a few clicks. I'm assuming they're not serving millions of users, that's when "scale" kicks in.

What does this have to do with containers? They just need a CDN in front of the site.

> being hugged to death

Whenever I read about this I can't help to think that people don't make a good use of a load balancer like HAProxy in front of their web-servers. Even if it's a single server behind the load balancer.

There's this sweet spot where your server is handling requests as fast as it can but if you cross above it everything halts. To make it simple, this usually happens when one or more hardware resources (CPU, RAM, disk) reaches 90% utilization.

So just do some basic benchmarks and find that sweet spot and put a limit in HAProxy. Say 100 concurrent connections for static assets and 50 concurrent connections for everything else.

All requests above the limit will go in a HAProxy queue but they should not stay there for long because the web-server is able to work at full speed without being overloaded.

PS: I have quite a dream. To write some fabulous blog post that reaches the HN front page and serve it from a raspberry pi device. Too bad I don't write at all.

This is informative, thank you. Any HAProxy-like software you recommend? Or is HAProxy the king?

You might be able to climb the ladder with a meta-post. Just do a write-up on how to host a front page post on a Raspberry Pi and tune the headline for the right audience. "This Site Can Handle 1,000,000 Connections on a Raspberry Pi" If you could collect data from the experiment and do a postmortem, that'd be great, too.

> postmortem

Hehe, why so negative?

I've actually dealt with this precise problem, HAProxy doesn't really solve the problem of an excess of traffic. At some point you'll start dropping connections or people will give up waiting.

What you really want is something more like Varnish, a pass through caching proxy that handles known requests from memory, and any novel requests go through.

I agree that Varnish might be a good addition but in the end it depends on the exact setup you're running and which hardware resource is used the most.

My opinion is that the CPU is usually the first to fall and you will not help it quite that much with an in-memory cache when the actual problem is the https overhead or the PHP/DB processing.

What you're describing is queueing, but the solution you're really looking for is back pressure. Any queue will eventually fill and when you fill it, things fall out. So rather than just dropping requests or letting the requests swamp your box, what you want to do is intelligently reject requests.

On the web, this means sending 503 responses. In a more intelligent client-server model, this might be the client doing an exponential back-off retry with jitter. A plain-old retry can swamp the server and just kill everything. An exponential back-off retry will allow the backend to achieve a level of service that isn't fast, but also isn't unavailable.

Hopefully in this case the haproxy is used intelligently and simply returns cached responses to GET requests without query strings. You don't want to hold onto connections or pass them off to the dynamic layer behind the proxy if you're overloaded. You can also respond to the client with a page with an auto-refresh, implementing an exponential back-off retry.

The setup I described will not get you from 10 requests/second to 1000. It will let you to make the most of your hardware though.

Reply on news.ycombinator.com