Mark Zuckerberg: "Ship the App" (2011) | Hacker News Reader

Mark Zuckerberg: "Ship the App" (2011)

techemails.com

90 pointsrurp77 comments

This mail is so very spot on. In large companies there are so many people trying very hard to sidetrack the company goals and would come up with a lot of excuses of why it should or could not be done. In a large company, adding a form field to an internal UI (yes internal UI which has much much less constraints than user facing) with an optional field takes so much back and forth and multiple committee meetings that the feature itself becomes secondary (Note that this doesn't happen if there is external pressure on the company for the feature, like competition or government mandates). I don't think this is a solvable problem easily other than keeping hiring very very selective and keeping the company as small as possible.

One example of this is the Bill Gates mail about MovieMaker where he points extremely obvious issues and in response none of the VPs says "This is unacceptable. We will take care of it". The response mail is just all of them saying not my scope.

I've definitely seen the same at other big companies.

I've wondered where the line is; what's the point at which the company grows too much and regresses to this sort of mess of needing to get bureaucratic consensus? Purely experientially, it seems to be at around the 40-50 people person mark.

I used to think it was an active founder/CEO that couldn't direct things anymore, but that happens at maybe 20-30 people, so that's not it.

What is the point where the company is eyed as stable for retirement goals? That’s where the subjective interests override the companies objectives and the growth of the company changes towards preservative rather than dangerous; wherein capital risk intrinsically is.

When you say 40-50 people, are we talking 40-50 employees, or 40-50 "director" level positions? Because 40-50 employees seems incredibly low for reaching bureaucratic mess levels. In my area, there're plenty of small-medium companies with 50-100+ employees that are essentially founder ran with maybe a few higher level leadership under them.

I'd even say if you're reaching bureaucratic mess at such low employee count, then it's strictly a company issue of adding too much management and too many silos so early.

40-50 total.

I agree it seems low, but for my sample size of ~3 companies it’s held.

i've worked at a 30-50 person company that was (is?) super inefficient. at some point there were more directors, managers and product people than actual engineers doing the work. i remember specifically that at one point there was a team of 4 people with 1 engineer and 3 different "stakeholders"!

Probably the product people would argue that these stakeholders need to be better product-managed, while you seem to be thinking it needs more engineers... and now we're off to the races...

I believe this is the app they were talking about. So after this email it still took some 8 months to ship Facebook Camera https://about.fb.com/news/2012/05/introducing-facebook-camer...

Thanks for finding this! I was wondering what 'app' they were talking about

That was a surprisingly direct and non-hostile directive. Wish my company communicated like this.

Yeah, I thought it was A pretty good example of how to say something strongly and directly without being hostile.

being politely hostile isn't not being hostile, it's just being hostile while maintaining decorum. there aren't meaningful consequences for a CEO breaking decorum in negotiations with those under them; maintaining it is just a nicety

while prior context is absent, at least in this email zuck isn't offering assistance or asking what he could do assist, it's just a politely-worded "get it done, fucker"

i find a servant leadership approach far more effective, and better able to acknowledge that said team problems were quite likely caused by previous "stop complaining and get it done" leadership whose only skill is cracking the whip

I mean, it’s a bit beyond polite I think, and in the lack of proper context we have, get it done seems like enough to me. (I’m assuming he is speaking to someone who is supposed to have a capacity for management and leadership)

If the PM needs resources, she should just ask for them, in this situation. If a PM needs handholding or niceties to do their job, they probably shouldn’t be a PM.

Making sure priorities are clearly aligned seems like an appropriate conversation.

On servant leadership, serving the greater good is the only legitimate form of leadership. But that doesn’t mean being servile to incompetence. That goes against the greater good. A leader defining what represents competence in this situation seems appropriate to me.

You can’t serve your way to success you are serving people with misaligned agendas or whom are not capable of doing their job. You serve the people who are serving the organisation. You don’t ask “what do you need” to a person who is apparently not even working on the same goals that the organisation needs to address. That is just fuelling the misguided path.

I feel this every time I read an internal Zuck email.

I hope to be this clear and concise some day.

Step 1: Be a multi-billionaire from Harvard

Step 2: Have absolute control of the company

damn thats crazy he was a multibillionaire before creating facebook?

No but he was when he wrote the memo they're commenting about.

Did you read the headline or the actual email? This is definitely not non-hostile. He flat out says "I don't care about the team"

no, he said it wasn't his priority. his business isn't in building teams, it's in advertising.

I don't generally like people like him, but I don't think he's done anything wrong here . he was actually very clear about what he wanted, a great feature in a boss.

He said he doesn’t care about fixing the team as a stand alone objective. Which is fine by me, because it’s probably a bullshit excuse for them not getting things done anyway.

"Separately, I just don't understand why this app is so hard to build."

That doesn’t sound hostile to me, it sounds like an invitation for somebody to tell him why it’s harder than he thinks, or to get on with the job.

Overall, the clarity and rationale of the email improves my impression of Zuckerberg.

I'm not a big fan of Facebook, but this sounds like he might be better to work with than the average tech industry executive.

Not knowing how they usually talk, I agree it's probably "non-hostile", but it might be a bit confrontational. And maybe that's appropriate for the situation.

In 2012, I created a photo app for the VK social network in 2 weeks. The app quickly gained 30k users and I was planning to adapt it for Facebook, but my partner disappeared. Later, it turned out that he sold the app to move to Thailand. If I had known back then, I would have sold the app to Mark Zuckerberg and become incredibly wealthy

Wow that's quite the story

Yeah, this wasn't the first missed project. The most interesting story came from my music app for iOS in 2010, which I developed for a year and which was in the top 10 music apps in the App Store. But the managers killed it because they didn't listen to advice and tried to sell it to footballers (the boss just had a sports channel)

I'm not close to these deals, but my understanding is you have to have some existing connection to the CEO or at least very senior VP's for it to be an actual reality. So I wouldn't feel too bad!

As I remember, there weren't good photo apps back then, and there were just a few good apps on Facebook, which was concerning, while Instagram was just becoming popular. Apparently, the managers were siphoning money from Facebook and sabotaging the work. My app was called InstaGo and offered filters like Instagram. I think it would have quickly become popular and reached the top and I would have met a lot of interesting people. Of course, the Facebook managers could have just copied it if there were no connections with the leadership. But I believe that in those days, it was still possible to reach out to someone in Facebook's leadership through Facebook

The more interesting question I have is how he knew the speed that IG user base was growing. The threat that Google may acquire IG when IG was barely a year old seemed quite paranoid, especially when Google had almost certainly ceded its Google+ effort by then.

If anything, the rumors in startup land at the time had Twitter as the likely buyer and IG was quickly integrating with them. It was a product that, to this day, never felt at home with the FB News Feed. Threads is the proof that it really should have been a Twitter product.

Zuck does make fair points, and his direct reports calling out "team issues" seems like excuses. If anything, the "efficient" mono corporate culture of Facebook makes launching a whole new product that isn't just a feature of a bigger product almost impossible. I was on a couple zero-to-one efforts that never made it past the prototype stage because it couldn't really find a home in the bigger orgs. Camera was no different.

> The more interesting question I have is how he knew the speed that IG user base was growing.

See the acquisition of Onavo by Meta/FB [1]. Onavo was a Spyware which allegedly provided the data to FB on which Apps they were loosing ground to. This very likely played a part in Meta/FB deciding to acquire Instagram and Whatsapp.

[1]: https://en.wikipedia.org/wiki/Onavo

Nice theory but Onavo was acquired post Instagram.

Why do you think FB only had Onavo's data after acquisition? See this article from when instagram acquired [1].

> Instagram has grown on the iOS platform and recently opened its community to Android users. The service now has 30 million users — up from 5 million in June — and the app is the sixth-most downloaded iOS application, according to Onavo's numbers

Quote from the article

[1]: https://www.washingtonpost.com/business/technology/with-inst...

Interesting to see these companies before they became "infinite money generators". I'm sure Zuck today could call upon 500 engineers if he had something that urgent he wanted done.

Nine women cannot give birth in one month.

They can, if you'd just use k8s in a multiple AWS availability zones. (Or was it lambdas? I get confused...)

That's because only one women can build a baby at a time. Whereas multiple people cann work on an app at the same time and ship it faster than one person.

The point you seem to be missing is that some work is quite simply not parallelizable, which is what people try to convey when talking about multiple women not being able to give birth to a single baby in less than 9 months.

You can staff up bigger teams to accomplish bigger goals, but this is far from linearly scalable, with different categories of problems showing up in larger organisations.

I have, many many times, seen a large organization "crash" a project, that is to say put a bunch of new developers into it. It almost never works. The first thing that happens is that everything stops, while the existing devs tell the new ones what is going on and what needs to be done.

The second reason it doesn't work is that splitting the app into separate parts that can be worked on in parallel, is essentially determining the software architecture. If you haven't sorted that out then the different parts will get built, not work together, and then there will be a sh*%storm of blaming each other for why they don't work together.

There are cases where new devs can help, but if you don't have the overall architecture sorted out yet, then they will not, and if you do then there is a finite pace at which new devs can be integrated and brought up to speed.

The old analogy to pregnancy is, in fact, spot on.

Fred wants a word: <https://en.wikipedia.org/wiki/The_Mythical_Man-Month>

Piling devs onto a project is counterproductive for fundamental communications reasons.

The three chief alternatives are:

- Highly modularise the project and treat those modules as independent projects.

- Launch multiple projects which compete against one another for ultimate launch. Again, treat those as independent projects.

- Buy competing projects. This is a variant of the second option. The competing projects are inherently independent.

Both approaches reduce communications overhead. The result isn't faster production, but rather risk mitigation and diversification over a larger set of investments. Both approaches also work best for an organisation which already has a large engineering capacity.

But Zuck has learnt his lesson and doesn't call for 500 engineers when he needs something quickly, e.g Thread was built with a much smaller team.

https://engineering.fb.com/2023/09/07/culture/threads-inside...

I see how the urgency of this email can be appropriate or even admirable, especially in social media where the fastest growing apps tend to dominate. I just wish they shared the same urgency when it comes to protecting their users—e.g., the latest case of password storage negligence that hit the news earlier this week.

https://news.ycombinator.com/item?id=41678840

I think it's interesting that Threads was built in only about 5 months, and from my understanding did well in the initial flood of traffic it received.

Threads was built intentionally with a very small team I think, for this exact reason.

I guess they had a mistake to learn from

Multiple comments here talking about how Zuck doesn't care about the team. That is not my read at all. He specifically says that he wants to both fix the team and ship the product. What he does have a problem with is the sole focus on the team (aka fixing the team as the milestone on its own) while the company is losing its competitive advantage. That feels like a pretty reasonable stance from a "wartime" CEO [0]. A healthy team in a failing company is not an outcome to strive for.

[0] https://a16z.com/peacetime-ceo-wartime-ceo/

Does anyone know to whom this email was directed? It's not on the techemails.com site, nor could I find it on the linked twitter thread, but I'm assuming it's not redacted in the discovery materials on the lawsuit.

So I’m more junior so I don’t understand why this letter is controversial. Is it not important to meet your business goals in a crunch and address systemic issues in some sort of post mortem or retro? I do understand that there exists “management debt” for intra team issues, similar to technical debt, but I don’t see how it’s problematic to ask to prioritize finishing a project that’s been signed and on for and deal with the issues later, especially with the state of the business at that time.

Maybe because all too often "management debt" get balanced on the backs of the engineers. We're always being led to one "don't care about the costs, get it done" situation to the next.

You can theoretically ship at any time. But it might be broken, slow, harmful, or fail some critical objective.

Team dysfunction is inseparable from tech dysfunction. Imagine one team owns three quarters of this service as another owns half of it. Or more likely, one owns a tenth of a problem space and another owns a fifth. Suddenly you need a fully realized solution. Who will step up? Well, you’re the one who actually needs it…but surely you don’t have to start from scratch? Surely…?

Then there’s just the fear of showing your ass. The day after it launches, who will be in hot water? Will it be you, because you agreed with the PM that your substandard code could ship? Why didn’t you flag the problem earlier? Get more help? Tell someone you weren’t already a superstar at this infra or that architecture?…

And then recall they did ship, it sucked, they bought Instagram. Then WhatsApp. And dozens of others

I guess the lesson is: yelling “go faster!” at a blocked-up engine is only gonna go so far. Mark’s (and many others’) move is to apply money to replace or get additional engines. Or roll up your sleeves and fix the org and culture…doesn’t have exactly “founder mode” romance to it tho, and there’s scarcely a harder task out there.

Multiple times he mentions the most obnoxious manager-level thing "I don't understand why this is so hard".

I don’t understand what’s so hard about it either. It seems like a pretty simple wrapper around infrastructure that Facebook already had.

That can be a good position to start from when trying to break a logjam.

Of course it must be coupled with genuine flexibility to change your mind when someone points out actual, real difficulties.

It's one of those things that sounds really obnoxious until you've been on the other side of the table. There's lots of things that can go wrong to make an easy problem hard, and some of them are only really detectable when you think about it from first principles. I've personally seen multiple designs that split a moderately hard problem into a set of extremely hard tasks, often in such a way that no individual task could be simplified on its own.

I am no fan of Zuckerberg nor a user of any of the things his company makes.

But he is an engineer, more so than most managers. He would know what goes into making an application like this, and so know that things are moving slower than they should.

I see nothing obnoxious about asking this very valid question under those circumstances.

chatgpt grant me an AI demon that can transform any manager saying this into a pixie trapped in a cage, attached to each of their ICs' heads for duration of the project, or something

He doesn't just say "why is this so hard?". He provides rationale for what he thinks is required and the effort it should take, posits some things he might have missed, and asks for more explanation. If you get a good-faith request like this from a technical manager, you should be able to point them at some docs or give a relatively concise response as to what they're missing. Honestly, you should have already headed this off, because the moment you realized you were in danger of missing deadlines, you should already be raising it as an issue and either explaining your plan or asking advice for how to get back on track.

If you're missing deadlines, you can't both complain a) when there's no accountability and the business is going poorly, and b) when your managers are trying to hold you accountable.

Ultimately your employer has to make money to pay you for your work. Keeping this in mind is one way to help you avoid yak-shaving, bike-shedding, and other weed-entering activities.

Depends on the business goals and why it's being perceived or presented as a crunch. A lot of it is just to make someone feel important rather than produce value in the market.

zuck is the founder of the company and has been at the helm since its inception

consider where that management debt may have come from, and whether accepting the current crunch will somehow alleviate it, or signal that crunch will solve all problems and should be used, likely more aggressively, in the future

>As soon as we launch a compelling product a lot of people will use ours more and future Instagram users will find no reason to use them.

It's so fun to watch after the history happened and see what worked and what didn't and how they ended up just having to pay $1 billion for their competitor. I'm dreaming of a world where companies can't just buy their monopoly, crazy I know.

We can't have a real understanding of it without knowing what "fix the team" means. If this team that is in a broken state is responsible for shipping an app, not fixing it means not shipping the app, that's seem very basic.

Starts with the business reasoning. Then, shouldn’t be hard. Finally, get it done.

That last line gets bodies moving

How so?

Probably the implicit threat.

Pretty easy to say when you got an app that prints money by stealing data.

I don't know what folks are finding inspiring about this. Mark rants about how they need to copy Instagram, and as history points out, they failed to do so and just ended up acquiring them. This is reactive management, the worst kind of leadership where quick, hasty decisions are made.

Was Mark already trying to court Instagram founders when he sent this email out?

"ship it now, it's easy to ship, I don't care about the team"

Heh and people wonder why we have such toxic managers in the valley.

Hard disagree, and I don't think your quote accurately summarizes what the email says. On the contrary, when I've seen teams in large companies focus on nebulous goals like "fixing the team", it usually ends up being a sad, sad exercise it futility. What I've seen be much more effective is to give teams a very clear, concise business objective that they can all drive towards.

Zuckerberg didn't say "go on a death march and burn the team out". In my view he gave clear instructions on which features should be paramount, which can be immensely helpful for a team building a product.

Giving teams clear objectives is a huge part of getting them to function. Most teams I’ve been on that experienced any kind of dysfunction were suffering from a lack of leadership. We needed something to work towards, not just work to do.

You clearly haven’t met many people who worked close to Mark or for a long time at Facebook. Mark is not an engineer and he does not understand team based development. The company he created is Kafkaesque with a bureaucracy which uses data to make personel decisions but in the most comically short-sighted ways. Like using LOC to justify compensation rather than impact. Mark created this idiot company that succeeds in spite of itself thanks to social dynamics and savvy acquisitions.

You can downvote me but it’s the reason they can’t make anything good themselves and why Mark’s email is actually hilarious. Google is captive to a similar handicap.

Yeah in my experience in large companies, nebulous goals just get gamed to the hilt by bad actors.

> Zuckerberg didn't say "go on a death march and burn the team out".

Is our standard going to be people like Zuck writing the literal equivalent? That's never going to happen

"I think we need to do both"

"Getting the team to a good state is not a milestone _by itself_ that I care about" [emphasis mine]

I don't like Mark, but it seems you read his note selectively.

Can you elaborate on why you feel this is a rant? I’m not a big Mark fan but generally it sounds like “hey you’re not hitting your goals and you claim it’s team issues and I need you balance the two” which seems like a reasonable ask especially when he explains why that’s the case?

Aside from llama I haven’t used any of what they’re “shipping” in a very, very long time. Is FB still a thing? Why?

FB is still a big thing in various circles.

There are so many other people than yourself.

Reply on news.ycombinator.com