# Paul Graham & Marc Andreessen Interview on if Marc Andreessen foresaw what was to follow from the Netscape browser: what he didn't forsee was the unlock of creativity. new things. new ideas. not for things that already existed plain text in all the protocols. so people could view source. that people could just do it was instrumental to the web's success at the time people thought binary. Text was a crazy idea optimization was neccesary. anybody sane thought transmission should be byte by byte. invention of the tag some people wanted the net "pure". just for the physcist (har har) image made newspaper, etc. possible netscape was "unbelievably good" at dealing with broken html added things that were "needed" javascript ssl started in 22. mistakes? would you do anything different? a) from what you know now they were getting qualitative data, but there was no instrumentation of their browser. they didn't know to instrument it. idea of "social" didn't exist. they would've done flock. integrated with your friends. payment didn't exist yet. no clear way to do it. now they would've put a "pay now" button. which would've changed what the internet is now today. would've been e-commerce way earlier, with more paid-content, and earlier. b) from what you knew then back then nobody knew each other. very sparse network. no information sharing. how did you grow so fast? both in revenue and people army of sales 1,500 in sales, out of 3,000 "building a company is like baking a cake" you have to add all these things at the right time. it was really hard to bake the cake in 3 minutes learned from Jim Clark: "engineer need to be in the heart of the company" ahead of his time company that does not continue to innovate start to go stale this is why you need the engineers at the core "move fast and break stuff" bible at the time was "crossing the chasm" wish could've spent more time on people it's incredibly hard to hire that fast. it takes time to integrate people you'd hire somebody from Sun, then because you are hiring so fast, sun people whould just pull more people from sun people from different companies brought their own culture. Netscape ended up with a lot of DNA from other companies, just because there were cliques as a result of the hiring process why did you start a VC rather another company? bigger impact experimented by being angel investors what have you learned? from 3 you founded, and those you funded the technical founders who become CEO are the best "engineer who becomes a entrepnreuer who becomes a CEO... there's something magical about that 3 distinct skillsets. what's the big surprise? CEO is a learnable skill. They are not just born. Learn sales. Learn how to manage. You can learn on the fly. you have to have want to learn, you can do it. getting trained, getting educated surround yourself with mentors, coaches, people who can help you. bulid an executive team around you. # James Lindenbaum, Heroku It's hard to build for developers, but it's very rewarding. Before they were consulting. Based in LA poka-yoke a lean manufucaturing idea. a thing that's in there that prevents you from making mistake. if you can screw up, how can you prevent people from making mistake. a lot of such devices are in Heroku's design. immediately started getting attention/traction when they launched "pull a paul graham". PG would talk to you about something, then bring up something complete related. he talked about Heroku, then out of nowhere, "you guys should kill Oracle" they had a lot of traction, but they felt they didn't have "product market fit" people weren't doing anything valuable they saw a segmentation problem people using the web tool were just playing around, trying it out serious people used git push they pivoted, and focused on those using git push / API hard decision at the time, because they were getting so excited, and getting traction. they ended up rebranding Heroku. Came up with "herokugarden", a user friendly version. Herokugarden is a friendly place you go to learn Rails. they focused on the developer. rebranded heroku itself to be more "hackerish" business people would go to the landing page, and be like, no, wtf is this? i don't understand a word things are moving to become services this is the new way applications will be built what's the next thing? polyglot programming. why they joined salesforce salesforce knows how to sell cloud to enterprise. salesforce allows them to make the biggest impact "best" for the mission they probably could make more money if they didn't sell heroku is independent they own everything could use salesforce's leverage they hired matz. admire him for focusing on developers matz's focus on developer happiness lead to rails, which lead to heroku. heritage it was a great honor for them this model is just taking off runtime in the middle, services out there call for action "start more cloud services. not for everybody, but it's worth it" understanding how to scale the team was the biggest challenge they reorganize the team every few months # Jim Goetz - Sequoia Capital Founder everybody started unknown. tenacious Idea Timing - is it too late A billion of smart phones in 18 month. I can get to 20B in 2020. half of who they back are immingrants # Matt Mullenweg First 100k are the hardest Wordpress was not written from sketch. From moveable type (which at the time had 90% marketshare) php & mysql: "it ran easily anywhere" at the very beginning, more developers than users (it's always like that) be your most passionate user to create a good UX, you need to be (or sleep with) a person that uses your product everyday. this is the only way you can feel the pain 1.0 is the loneliest number check this out: ma.tt/2010/11/one-point-oh wordpress was one of the first to adopt webstandards and css because of that, they got a lot of early adopters who cared about it it's important to get it out to real users there's nothing better than the crucible of real usage "remember the silent majority" internet is almost designed to reward the most vocal, and those looking for a debate famous five-minute-install it was not famous. "I made it up" it was easy at all, but compared to competitors, it was {marginal advantage was enough} "there's nothing quite more permanent than a temporary business model" "you need a better taglne" You have to set the bar that everybody follows. That is part of being your passionate user. support@ should be everybody@ do this for as long as you can possibly keep up version 1.2, 822 downloads a day version 2.0 5,115 downloads a day After CNet, Started automatic. At first, worked on Akismet. Fight spam. Wasn't sure if there was space for a blogging company. There was MySpace, Google Blog, AOL They originally had revenue from hosting companies. When starting wordpress.com, they were going to lose that revenue. They weren't sure if there was money to be made. Passionate people: they signed up for a crappy product # Mark Pincus The earlier years seem more exciting and romantic. In retrospect. exciting, but few and far in between. feels like many things could kill it tribe.net didn't fail fast. fail fast if you can "what do i want to do for the next 10,20 years?" i want to do consumer product. it's hard to fail fast. after you found out, it's Series B, 30 people company how about i build a large audience first, then try to go from there? was looking at acquiring CNet then facebook opened up their platform. it was a mind boggling opportunity you have all these audience ready for your product decide what you want. then take your risks around that biggest lesson from tribe value engineering time. brutal on testing. does this thing we are doing help? where's the heat? where's the candy? what is it that we absolutely know people wanted? we try to figure out what people want poer game: 80% they clicked on fold, they'd stay in for the next game. what can we give them to do while they are waiting? WHATEVER we needed to do to spice up our poker game. strippers. dancins gnmoes. we try to find metrics that would repeat i feel like we are launching a new startup every week each game is like a startup. many things has to come together for something to work people must really want something. must tune into that. what is the essence? what's the unique value position? if you describe it in 5 words, if you send those 5 words, would your mother want it? Zynga wants to make games instantly social, without you doing any work lots and lots of metrics in the backend. how good are we at socialness? how quickly does a user find friends? how engaged is the user with friends? are there features we can change to amp up the socialness lesson learned in the previous startups. "i had a lot of scars from tribe" what would Pincus do again? build a house you would want to live in. do you want to work on this for the next 10 years? if you have convicition, it's much easier to press on *go all in. give everything you have. don't figure out an exit path. "why am I a serial entrepnerues: I kept failing. I couldn't build a sustainable company" surprises? he never had this level of product success before zynga. march 7, 2008 first day a virtual good was sold. wow. why would people pay for poker chips they can't cash out. but they would! after tribe, he had a lot of coaching it's good to have somebody that tells you: you are not thinking big enough Jeff Bezo after 300 people, your company is going to be different. the people you are hiring are not going to remember the founding. they don't have the muscle memory for startup, and they will be risk adverse. your job as a CEO to the end of your days, will be to get them not to be risk-adverse. Bezo was right. But he didn't give an answer. Mark Z singular focus. biggest mistake first founders make once you have a product and a biz plan, most ppl prioritize over the wrong thing prioritize keeping total control of your destiny they want to look for people to invest, to see if they have a good idea what is your big macro goal? write it down. connect the dots. if you don't have the goal, you'll be making lots of compromises. if you don't have the goal clear your mind, you won't be on that path # Paul Graham Office Hour When reading about YC applications: so many people the desire precede idea most ideas are derivative best ideas are from people trying to solve problems look for problems. look for something broken. there's where good ideas come from. identify the initial users that are having the most pain users that are most desperate if it works, you are gonna take 5 years to pull it through. you should do something that matters and you can do 5 years for. the oracle speaks: for dating web-sites, build a matchmaking tool, for matchmakers. Cultivate busy bodies. You can break the chicken and egg problem. The matchmaker would be like a dating website with population of N=1. # Mark Zuckerberg Most companies mess up by being too slow, by being to precise. you want to make mistake on both sides. mess up by beign both too fast, and too slow, so you are in the middle. in order to find market equilibirum, you have to make mistakes on both sides the biggest risk is not taking any risks in a world thats changing so fast, you WILL fail if you don't do stuff, and making changes this is the biggest risk one of the defining characterstic of facebook is that they are an extremely technical company when you have technical people who think about these problem and think about the product all throughout the stack, you have something special people making the product decisions also know the technical tradeoffs. that's special. the direction is social. everything would be better if your friends were there. when setting up company "we literally knew nothing" made lots of mistakes. kept iterating. we didn't go into it to build a company. a company is the best way to align talented people to make the changes you want in the world. they get rewarded both financially, and fulfilling the mission the way to scale is to decompose the problem. 50~3000 is more or less the same for Mark Z. there are some complications, politics, etc. "friends being there" is a feature. growth, scaling, and the company competence for that are designed around that core feature. growth for facebook is something deeper. deeper for facebook than other companies they have a growth team. once you have a product, and ready for growth, should centralize that competence. once you have a core product, help your user to take advantage of your product as easily as possible for facebook "10" is the magic number. If you have 10 friends, you have enough stuff to keep coming back for so they do things to focus on you getting 10 friends on selling companies ultimately, it's about what you want to do. "we really have no idea what we were doing" AS CEO, build the product. Recruit people to build the product. 1000 things going on at the same time, but only a couple things that matters. you can be so bad at so many things, as long as you get the things that matter right stays focus on customer. being unique and valuable. you are gonna make a lot of mistakes. it'll be ok. people only remember what is good. JL: what do you do right to win? the awesomest thing here is you can just try it. and see if it works. how MZ works: you really need to start with a passion. you want this thing to exist. people want it. you are just pulled along. do something that you like. the past 5 years is getting people to get connected on the internet the next 5 years is going to be about what people build with this connectedness you can build on the social graph, or you don't if you don't you'll die. the graph is not about the facebook app. like when internet was wiring up if you worked on CD-ROM, you die if you worked on Internet, you live If you are a beginner and you don't know anything, the valley is a good place to start. there are things you can't learn on your own, but if you are here you can learn from people "If I am starting now, I'd probably not bother" why not? short-term focus here. people want to start a company, just to start a company. people don't commit to doing things the average time spent in a job in seattle is twice in silicon valley There's an advantage to doing things different from everyone else. There's probably a reason why people aren't doing it the same way, but You could fail miserably, but that's how you really win. Maybe there's a way to express this mathematically, but... don't know? He asks his employee "what are you doing differently from others" # Jason Cohen, Palantir You can pay somebody to get work done, but you can't pay somebody to drink red bull and work for months abd months. Do what you love. # Max Levchin Topic: Founder & Co-Founder 4 failed startups. then paypal, then another "I am one of those crazy guys who can't get enough of this stuff" the job of a co-founder is not like "man... we are fucked", and he's like "ya, we are fucked" at the lowest point, the cofounder's job is to provide support when you are most beat getting emotional and practical support respect if you feel it's not going to work... break up. if you ever have doubt that this person could be your co-founder, run away. It would never work. how to fire someone, and it feels free for you and the person you are firing "i'd like to ask you to resign" it works. every time. doesn't know why it works, but it makes it better. don't confuse "hard" and "valuable" if you have a really hard problem to solve, ask if it's valuable for anyone. if the answer is no, don't bother "once of my more honest investors pulled me aside..." "the underpant gnome business model" collect underpants. ...? Profit! something that plaques us is fear of failure, and lack of objective nothing matters except passion for product all other emotions, like fear and embarrasment, are extraneous. starting a company requires an odd emotional detachment # Ron Conway Topic: The Defining Entrepreneur "In your heads is the future of innovation" The traits and lessons "Ronco" (what PG calls Ron Conway) learned from the defining entrepreneurs of the last 20 years. "Think Big. As ambitious as you want to be. Go for it." usually a product visionary. owns the mind of your customers. decisiveness make a decision and move forward clear vision, and make sure the entire team shares the vision listens well, but strong welled moving fast 24/7 work ethic be relentless. be scrappy. always keep a bootstrap mentality lead by example execution is everything make mistakes, but learn from it trust your gut your reputation is your biggest asset. don't compromise your reputation as you build your business. larry & sergei were sponges of learning about business # Drew Houston Everything big starts small. Everyone starts out clueless. There are benefits from never having done it before. Take on more than you're "ready" for (and get used to that feeling) the fastest way to learn about startups is to join one put yourself in a position where you are going to be pulled to learn surround yourself with kindred spirits: "you are the average of your 5 closest friends" If things work out, you can actually change the world.