> People in Silicon Valley are dicks
> Hire those bothered by suck
> [random butt statue]
This talk got restructured at least twice this week before I gave it, but I should have reworded those phrases, too.
Maybe we should look on this as an opportunity to alter our vocabulary. I bet most of the people on this board are like me -- when they were twelve, they had a vocabulary like a college graduate. But somehow as adults we decided it was cool to talk like twelve-year-olds.
The story is a few thousand years old http://en.wikipedia.org/wiki/Venus_Callipyge#Interpretations The "objectification" might be considered sexist now, but I doubt the original story meant to belittle the women in any way.
If I had to bet on one of those choices, I'd say the culture is the cause, not the result. But, if I had another choice, I'd say probably neither. Because "success" is a complex formula containing many components, one of which is no doubt culture.
Github has an awesome culture because they want to have an awesome culture. It is not the cause nor the result of their success. I think the core value from which all other things stem is simply, to build. IMO that's a value, more than a characteristic, of their culture.
The fact that their company is also successful was no doubt helped by these things (and certainly validates their approach to running a company), but, as Holman hints at in the slides, there is no recipe for building great products -- your best bet is to just build, and keep building.
They continued to be successful by creating a culture that attracts good people who continue to build good products.
http://www.sfgate.com/technology/article/Andreessen-Horowitz...
[1] http://cofounder.tv/videos/github-founder-bootstrapping-vent...
That's gold.
A great idea killed by a BigCo bureaucracy is like a talented athlete who is injured before his/her pro career even starts. All the talent in the world will never put such a person in the record books or hall of fame.
No matter how good it could've been, it still wasn't shipped.
In an older job, I was constantly battling my program manager's desire to have everybody "come in on time". It was a great offense to her if somebody came in so much as five minutes late. Despite being the managing developer for the team, and hence, ultimately responsible for all the engineering work that happened, I refused to care, so long as the work was getting done. I had a great team of talented guys that were always down to knock out a problem, whatever the problem was.
My boss wouldn't see the hours they put in working from home ("How do I know they're working if I can't see them!?!?"), or the hours they stayed late when the work wasn't done.
I tried the age old arguments "So long as the work is getting done..." or "They can come and go whenever they want so long as I'm meeting deadlines..." etc., but none of it flew. I regret not trying harder to change the culture before ultimately giving up and going somewhere that 'got it'. As a result, I have less responsibility, work from home, work more than I used to, and am happier to do so. I also can't imagine giving up the team that I have now for any reason, and I honestly think they feel similarly.
I usually poo-poo all over 'company culture' lectures, but this one completely hits the nail on the head as far as I'm concerned.
Anyone can put on a suit and tie, but the ability and desire to fix problems isn't inherent in everyone. I've been at government contracts where they overvalue the employees who dress well and don't value those that look them straight in the eye, tell the truth and deliver in a timely manner.
Government work is hard for the reasons you mentioned, and also because generally, contracting pays well enough to price you out of other markets.
My exit strategy was to find a government vendor (e.g., that makes products for government / enterprise consumption) vs. a government contractor. The services side of the house is still pretty much a suit-and-tie affair, though we pride ourselves on being much more meritocratic than our customers, but on the engineering side, pretty much anything goes.
The one thing I will say about gov contracting is that as a result, I interfaced with much more than would have been in my normal comfort zone, and I came away from it MUCH more knowledgeable than I would have had I not done it at all, and the work was very rewarding. My advice is to hold out for the good parts as long as you can while quietly looking for a way out that seems fulfilling.
I do have some questions though. What is Github's structure like? Is it completely flat? Are there project managers? How do you 'manage' remote workers? I feel like there has to be a little structure to the chaos, or at least some techniques and tools that make things work smoothly, especially with a company that's growing so fast.
We of course have "primary responsibility people" who are sort of shepherds of various features, and inevitably leaders emerge from among the pack. But we don't have anyone who simply "manages" things.
Our structure is able to withstand this due to the way we structure our internal teams. We're constantly tweaking how we do that, but by keeping the teams small, strategic decisions close to the people who are actually executing them, and retaining intense focus on one piece of the GitHub Product™ (which could be anything from a feature to "performance" to "making GitHubbers' lives easier" to "making us more money"), it makes things run smoothly without some sort of overlord forcing it.
Update: This post suffers from a cause-and-effect fallacy, and hindsight bias. More: http://times.jayliew.com/2012/10/12/cause-and-effect-fallacy...
So many developers are falling for this. Wake the F- up.
The irony: Zach doesn't realize he's talking like a pretentious Steve Jobs. Until you can prove cause and effect, please stop tricking developers. Yo.
I had a hard time taking Mr Holman's work seriously at first because of his choice of self portraits. They scream "brogrammer" to me.
I'm a convert now, though
Edit: I'm an idiot.
Worry more about *building* the damn thing
Worry less *about* the damn thing
There are so many things that are admirable and awe-inspiring about Github, and how they've gone about building their business. From my perspective, all of these things are a byproduct of one core value: building.
"I hate brogrammers" - but if I was asked to name a famous one, Zach be it. Excessive swearing in talks, "people on the internet are dicks", "OAuth will murder your children". If Zach Holman was holding a work barbecue, would people take their children?
Zach, this is the first of your slideshows that has left me with a lifted feeling, the feeling of reassurance that not all popular hackers are dicks, narrow-minded, or naive. Thank you!
I should probably be less judgmental, but I'm so tired of the facebook-rockstar-ninja vocabulary. Glad to see I'm not alone.
I'm in a BigCo and every time we acquire a LittleCo I get massive culture envy. I think you nailed how great product happens.
Isn't pivot is a synthetic replacement for the essence of 'happy' accident?