Hacker News new | past | comments | ask | show | jobs | submit login
Reflecting on 3 Years At Facebook (facebook.com)
87 points by arjunb on Aug 11, 2010 | hide | past | favorite | 44 comments



I despise posts like these. I hate being so negative about it, but it doesn't say anything. "Facebook has smart employees. Facebook wants you to learn new things. Facebook has impact." None of it is novel or interesting.

I wish there were more of a culture of insider writing while the events were going on. When Facebook pushed Questions and then retracted it, I'd love to see a blog post about "Yeah, we pushed Questions because it rocked but once we shipped it everything exploded and shit went crazy. Here are the details and what we learned from it. [...]" That's far more interesting to me, and really compelling to someone looking to become a Facebook engineer.

Unfortunately, once you grow to a certain level, those things are no longer kosher. Plain talk is discouraged in an era of Corporate Speak. It's unfortunate.

But cheers on this guy's three years though.


At least he didn't mention how great the food was. That's a step up from every "I love being at Google" post ever.


I love the food at Facebook. People underestimate how important good food is.


Yes, I'm sure it's nice. But it doesn't make for very interesting reading material. It's like knowing that your office has view.

It's interesting to read about what technologies companies use, how they use them etc. Not about things you could see if you flicked over to Martha Stewart.


But not to everyone. One of the benefits of working in SoMa is that people who care about good food can get it, and people who just want to grab a burrito from the taco truck can do that too.


What's wrong with a truck? Would you not consider that an approach to mvp cooking? In Austin (for which Anthony Bourdain complimented us about), the trailor-park restaurants expressed magnificent culinary arts. In fact I know one that started as a trailer until their demand proved significant enough for a bank loan to expand. Considering the failure rate for restaurants is not too dissimilar from that of startups, I think it's perfectly acceptable to start small and grow until the pompous diners accept you... at which point, you've sold out.


I get plenty of great food here in Italy, but I would absolutely kill for a decent burrito. In some dream future where I've made a ton of money and decided to continue living in Italy, I would open a good Mexican restaurant here. It would probably lose money, but I would not care. I would have good Mexican food to eat.


Isn't that the same thing?


Do you agree with the OP about the values of FB?


I'd tell you about the food at Apple, but that information is strictly confidential.


I agree, and I work at google.


There's also the aspect of plain talk about concrete lessons learned being valuable to competitors in a way that feel-good self promotion isn't. While it'd be nice to be able to learn from peoples' mistakes the first time those mistakes are ever made, it's saving a lot of money and effort that your competitors will have to make to learn the same lessons.

Once the knowledge of what works and what doesn't is wide spread enough, it's no longer an issue, and you can write a feel good fluff piece about it without getting into the gritty details, and get most all of the beneficial karma from fans and community, without giving away too many tips to rivals. I'd argue that the reason it's not kosher to give away secrets of how you did something at a big place is that once you get big enough, there's always someone around who can argue that the secrets are inestimably valuable, and the benefits in the form of developing community are paltry by comparison.


I'd rather be making those mistakes and pushing the industry forward than being the one taking the scraps from someone else's blog. You may mention part of the solution in your post, but the solution that takes one sentence to detail actually may have stemmed from weeks and months of detailed, considered thought.

Sure, the competitor may then bypass all your discussions and go straight to the "solution", but I'd argue the discussions and considered thought puts your company in a far better position to out-maneuver the competitor in the future for the very reason that you've been thinking about the problem domain much longer, and in a much broader sense. Getting to the result without the understanding doesn't yield long-term success.


I'm not sure what this means: "When Facebook pushed Questions and then retracted it"

We didn't ship and then retract. We just started with an extremely small seed community and are rolling it out very slowly over time.


I'm just referring towards my experience: I ran it for about a week, saw that the majority of it was broken or the caching was completely insane, and then I was dropped from Questions. I figured it was a botched launch. No disrespect intended; I just had no visibility on any of the background with it.


@blakeross: SF. You can ping me as @holman over Twitter if you want to take this offline and not clutter up HN as much. ;)


This also happened to a friend of mine, and he thought Facebook pulled Questions until I showed him they were still up for me.


Where are you located?


"Move fast and break things" is not the same ol'. I work at Facebook, and we really, really mean it, for better and for worse.


Compared to "I adore working at Google", you have to agree posts such as these are a rarity. I found the post informative, yes, perhaps a bit PR-looking, but it was a great read just a day after pg's article on hacker culture.


Interesting, but hard to consider a "look back" posted on an official Facebook blog as anything but necessarily whitewashed. (Not knocking.)


Agreed. Things are probably not as ideal as that post makes them sound.

What facebook is essentially saying with this post is: "We're an engineer/programmer-driven company. If you're a good engineer, you'd like working here."

The hoped-for effect is probably to continue attracting good engineers. Many companies that aren't engineer-centric probably wouldn't even know how to say this in their official blog, much less actually bother to say it. So the fact they're saying it still carries some weight.


The "break things" part of "move fast and break things" is something Google could learn a bit from. (Speaking as a Googler.)

We're great at adding 9s of reliability, not as great at sacrificing reliability for progress. You can only move so fast when everything must work.

This is the liability I guess of having a huge launch audience for even minor products.


I can't think of any Google product which is as frivolous as FB. Maybe buzz and wave (RIP).

People would not take down-time in gmail, analytics, calendar, search, maps, et al lightly.


I work near the people that did goo.gl, and even though it has higher uptime and lower latency than any other url shortener, there's pressure on them to make it even faster and more reliable. We definitely have an institutional mentality that speed and reliability are the best features, and every other feature is secondary. (Not saying this is a bad thing, it just biases us towards certain types of products.)


For most "panda" tasks, isn't that the case? When people use a product to do something, they evaluate it based on how well it helps them accomplish that. Speed and reliability are important features there.

This model completely falls down for "lobster" tasks, which may be why Google sucks at social. But a URL-shortener, along with most of Google's products, is a panda task. I don't think that the emphasis on speed & reliability is misplaced there.


Sorry, can you explain what constitutes a "panda" or a "lobster" task?


http://ifindkarma.posterous.com/pandas-and-lobsters-why-goog...

Was all over Hacker News for a while, with several discussion threads referring to it afterwards.


I can. It's very easy.

Everything Google makes apart from Google search.

The rest of Google is minor and tiny compared to Facebook.

They have one seriously profitable product. At the moment, the rest of it is all fluff.


I wouldn't say Android is fluff.


Agreed. I wouldn't say Gmail is fluff, either. Of course, he's equating fluffiness with non-profitability, but I think that's wrong. I doubt he'd be willing to stand up in front of a roomful of geeks at a conference and call Twitter fluff.


Speaking as a developer of Facebook apps, Facebook truly does like to break things, and it is incredibly annoying.

Google doesn't need any of that.


Keep it that way. I run Apps, Voice, Adwords, Analytics, and soon Checkout and the inference API.

I absolutely don't want any of it fucked with.


Wow, Facebook's site says Facebook is a good place to work?


It's like reading one of those breathless self-promotional google or palantir blog posts.


One question I would have is, "how are things different now?"

A lot could look similar on the face of it, but the company has grown so much - in users, mission, employees - that that same experience may not apply at present, even if highly recognizable components remain.


I just passed the 3 year mark not too long ago, I wish I felt like that about my time where I am. What an experience that must be to come in at that time in Facebook and come from an amazing place such as Google too.


A lot of staring at screens and eating free food.

There's more to life...

Oh wait, you get to wear a sombrero.


A sombrero might have changed my last 3 years completely. What I mean though is that I wish I could have done more than I have done in the last 3 years, I don't feel like I've moved forward much.


You know what you need to do. You've got to push yourself forward! Try not to lament about the past, just concentrate on making your future better.


..by purchasing a sombrero at once.


awesome post, until he calls the folks he manages "rockstars." can we please all just move past that ridiculous cliche??


I like it more than "hacker."


For me this was the first (real/substantial?) blog post i've read on Facebook. Just throwing that out there.




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

Search: