One Minute Before Three O'Clock

The following started as an account of my Dev Bootcamp experience, and evolved into a collection of essays about learning, web development, and more.

[dnlgmb]

DBC F12 Demo Day

This past Thursday December 6th, 2012 ten teams showed off their elegantly crafted final projects. After eight weeks of intense learning and two weeks of development the verdict is in: our formal learning institutions are wildly inefficient and alternatives like Dev Bootcamp are changing how we do learning. Enjoy the show:

Instatapes

http://www.instatap.es

Hang Your Boots

http://hangyourboots.herokuapp.com/

driiive

http://frozen-garden-5238.herokuapp.com/

Filters.io

http://filters.io

Speech Hacker

http://speech-hacker.herokuapp.com/

ApprenticeHour

http://apprenticehour.com

All Day Fantasy

http://alldayfantasy.herokuapp.com

wi-chai

http://wi-chai.com/

verbza

http://verbza.herokuapp.com

fabawag

http://fabawag-demo.herokuapp.com

[dnlgmb]

Test for Best Career Path Alternative

Not too long ago, I was having a conversation with long-time friend Matt Harris about what factors should influence career decisions. He had just pivoted back into finance from tech and I was pivoting from finance into tech, so we talked for a while and came up with the following litmus test.

Is this the path that...

...maximizes your interaction with quality people?

...unlocks the most quality doors?

...maximizes your desired learning?

...offers stimulating work?

...fits with your financial goals?

This framework assumes that you have some idea of what drives you as a person because without a clear set of guiding principles you'll just follow your life's momentum without really effecting it.

There are two primary themes here, growth and flexibility. Growth refers to professional growth and, at a deeper level, personal growth. Flexibility stems from the fact that career decisions are often made without access to perfect information, so ideally you want to stay agile.

Of course, there will be plenty of times when constraints will drive new specifications into the decision. But the key is to develop a reliable framework for personal growth and customize it as necessary. And when all else fails, you can always ask yourself the most fundamental career choice question: will I regret not taking this opportunity?

Recommended Reading

Ben Franklin: America's Original Entrepreneur

[bootedup-deactivated20131225]

Introducing: Dinner-Table Tennis

A bootcamp-style education cultivates creative and innovative solutions to all sorts of problems. Four-dozen brilliant student minds collaborating. And now I present to you: dinner-table tennis. 

Jonathan de Pepinski - the reigning champion, hailing from Montreal, Quebec, CA. Also known as “Jauny”. 

Note his kung-fu-like Karate-kid-style forehand. Brilliant form. 

[flavorite-blog]

TL;DR

When a user creates an account, she provides an email and a password. The email is stored in clear, but the password is converted into a encrypted before being stored. When a user logs in, the password ‘guess’is encrypted using the same technique and is then compared to the encrypted...

[dnlgmb]

The Value of DBC (and perhaps YC?)

I wrote about the frustration safety valve being the most valuable feature of DBC, but in fact that is only the surface of a much deeper, subtler truth. In a way -a critical way- DBC is similar to YC. I've heard it said that the most valuable thing YC offers founders is its network, and part of the reason for this apparently is because it provides a social safety net for when things go wrong (a la frustration safety valve) and because it gives you access to other people with relatable experiences, which reduces the moral weight of starting a company.

One of the main reasons starting a company is so hard is the constant disparity between the reality you are trying to create and the reality that actually exists. That is what Paul Graham refers to as moral weight. As a founder, you are constantly interacting with people who are skeptical of your vision, you are constantly dealing with facts that tell your logical self, the part of your brain doing life risk management, that chances are you'll fail; there is constant doubt and questioning from so many sources, circumstances, people, incumbents, media, friends and family. 

That is why programs like YC and DBC are so valuable. You are trying to change reality. You are trying to change the way things are. Whether you're trying to improve people's lives (or your own) by creating a new product, or you are trying to improve yourself by learning web development, you are staring reality in the face and saying, "I'm not satisfied with the way things are, and I'm gonna take things into my own hands. I'm gonna catalyze change."

That takes courage. Lots of courage. It takes courage to stare into a friend's eyes, watch him be skeptical of your vision, and still pursue that vision. It takes courage to tell someone you're giving up a six figure salary to pursue extreme uncertainty. It takes courage to tell someone that you love that you cannot be with them at the moment because there is something greater than anything else trying to get out of you. It takes courage to fail repeatedly in your attempts to bring this out and still continue trying.

At DBC, and presumably at YC, you are surrounded by people who have made the courageous decision to foregoe short-term comfort in exchange for an opportunity; the opportunity to live the life they want to live, regardless of the reality around them. You are tuned into a channel that's constantly streaming inspiration. And that inspiration supercharges you, it gives you strength. It is a fountain of courage and you are plugged into it, feeding it and feeding from it.

Can you imagine, if the whole education system worked this way? the things we could build? the knowledge we could discover? the lives we could live?

[dnlgmb]

How I Named My Blog

A few months ago, when I officially made the decision to start triathlon training, I came across the book Iron War by Matt Fitzgerald. To put it in perspective, that book is for triathlon what Michael Lewis' The New New Thing is for tech. You can get a good idea of the culture around triathlon by reading it.

Essentially Iron War refers to the 1989 Ironman World Championship in Kona, Hawaii, where the world's two greatest triathletes, Dave Scott and Mark Allen, raced shoulder-to-shoulder for eight hours until at one minute before three o'clock, with roughly one mile to go, after having swam 2.4 miles, biked 112 miles, and ran 25 miles, Mark Allen increased his speed slightly, which was already sub-six minute per mile, and Dave Scott's body could simply not respond. That time, one minute before three o'clock, the last moment Dave Scott and Mark Allen were shoulder-to-shoulder is so much bigger than one triathlon, even if it is the Ironman WC. It represents a relentless pursuit for self-actualization that is tremendously inspiring, yet so rare.

So why would I name my tech blog based on triathlon history? Because the experience of racing triathlons is not all that different from what we're trying to accomplish at Dev Bootcamp. In fact, it's not that different from life in general. From the book:

"There is no single characteristic that defines human nature or the human condition, but it is tempting to formulate such definitions all the same. Whichever characteristic stands out as the true essence of our condition or nature for any given person depends on his biases. My biases are such that the voluntary suffering of triathlon, borne in pursuit of the triumph of spirit over flesh, seems the most potent metaphor for our nature and condition.

"Our earthly lives are doomed. Our lives are, in the long run— the billion-year perspective— utterly meaningless. And in the meantime they are a constant struggle. It takes a lot of courage just to make it from one end to the other. Yet who among us tries merely to make it? Somehow we muster the courage and spirit not only to survive, and bear the suffering required for that, but to create meaning, to pursue greatness, to put on a show, and we willingly bear incalculable additional suffering for the sake of these nonmandatory joys. To paraphrase Dave Scott, something within us just can’t resist trying to “see what we can do” with these bodies and minds while we have them, no matter the price.

"Human beings are the creatures that try harder and never stop trying. We try when it’s hopeless. We try when it doesn’t make any sense. And when we succeed in trying harder than we have ever tried before, we try harder still. We are Cinderella, and Rocky, and the Little Engine That Could. We are beautiful in this way. Heartbreakingly beautiful.

"I love humanity most when I experience something that makes me feel a sense of solidarity with my human comrades as fellow doomed “tri-ers” (or is it triathletes?). Nothing makes me feel this sense more acutely than watching someone like Peter Reid run through illimitable misery for a reward that is only imagined to matter.

"And no experience of this kind affects me more powerfully than Iron War, the race where two men tried as hard as anyone has ever tried for anything."

-Matt Fitzgerald, Iron War: Dave Scott, Mark Allen, and the Greatest Race Ever Run

[dnlgmb]

it should be_ok (to share asynchronously)

It would be bad to filter out asynchronous learners from Dev Bootcamp (that is, people who will spend more time learning on their own and less time pairing).

There seems to be this fallacy floating around DBC right now that suggests that in order for this to be the most effective learning environment it can be, people need to pair program as much as possible. False. My purpose here is not to list those necessary things but rather to dispel the myth that we need to pair 100% of the time (or even often) to effect the most learning.

A quick story will illustrate the point best. Sometime last week, my roommates and I were sitting in our living room studying independently. We have developed this habit of going home around 11pm only to get home and continue working for a couple of hours more. I'm not sure how it happened, but for some reason Alan and I were discussing Rails and he referenced a lecture he had been watching. Now you have to know, Alan is famous for finding random pieces of quality content, so I've learned to trust the content he uses. The piece he was currently consuming was the SaaS Rails course from Berkley, and once I started watching it I was hooked. It does a great job of providing a foundation from which to map out the Rails knowledge we're all currently seeking, and it does so in a relatively short amount of time. Good, efficient learning.

The next day I ran into Jeffrey in the DBC kitchen, and he mentioned his frustration with the lack of visual content with which to study Rails. I figured the SaaS course did a good job of helping you visualize how Rails works, so I told him about it and encouraged him to check it out. Once Jeffrey realized the SaaS resource was great, he was inclined to share it with anyone else he spoke to. Perhaps the reason he was more inclined to widely spread it was that his level of frustration with other resources was much higher than mine, or perhaps it's something more subtle. Irrelevant. The point is that Alan found something that fit his learning style and shared it with me in passing. I happened to run into Jeffrey and make the connection between his frustration and this resource. And Jeffrey noticed the resource could help many more people and decided to distribute it en masse.

Here's the kicker: would this still have taken place if we were led to believe that we should always be pairing? Probably not. My point here is not to suggest that DBC staff have forced us to pair and only pair. Far from it. But lately there has been an air of "thou shalt pair as much as possible to increase the collective learning," and this view is inaccurate at best.

Sharing knowledge occurs from the natural diffusive properties of a learning environment and is independent of pair programming. Yes, pair programming creates diffusive circumstances, but so does learning alone at DBC. Even if a student learns alone, she will at some point pair program, or talk to someone during a meal, or offer a recommendation during a late-night conversation.

The key is not forcing, or guilting, or even encouraging people to pair program. It is creating an environment where the friction of social interaction has been reduced, which DBC achieves superbly, and where people are encouraged to follow their learning intuition - if you want to pair, pair; if you want to code alone, code alone; if you want to watch lectures, watch lectures; if you want to read documentation, read documentation.

If the goal is to diffuse knowledge through pair programming, then by all means we should be pairing all the time. But if the goal is to diffuse knowledge period, then perhaps the spectrum of diffusive processes is much wider.

Feedback Request: I love Dev Bootcamp and think it is a phenomenal place. My intention here is to help make Dev Bootcamp even better than it already is in the hopes that it will continue to grow and improve well into the future. If you think that intention doesn't come across drop me a line. I'd love to hear your feedback.

[alancohen-deactivated20160627]

Convention over configuration—or is it confusion over configuration?

I’ve started venturing into Ruby on Rails territory this week and I will admit that it has been a bit overwhelming. The power of Rails comes in its ability to let you go from ideation to prototype in a matter of hours. It does...

[dnlgmb]

To Remember: Controllers (plural) & Model (singular)

Think of an airport. The air traffic controllers direct airplane traffic, which means many planes pass through the hands of the air traffic controllers.

On the other hand, each plane's main concern is its own. It needs to take off, transport its passengers, and land. It communicates with air traffic controllers only to ensure it can do these three things safely. Its purpose is individual in nature, whereas the purpose of air traffic controllers is to coordinate many airplanes.

If you think of Rails controllers and models this way, it's fairly straight forward to remember the naming convention. Controllers handle many model instances, hence a controller's name is plural. A model represents an individual record, hence a model's name is singular.

[triptagupta]

“For the past few months, I’ve been exploring software design patterns, both at their roots and in their modern implementations. I’ve been exploring new languages and deepening my understanding of the languages that I already use. Man can I ever tell ya, it’s been exhilarating.

I started reading...

[dnlgmb]

Dev Bootcamp Trumps Learning Alone

They key ingredient of DBC is not the highly qualified staff, nor is it the inspiring visionary who leads it; it isn't the highly motivated students, nor the cutting-edge technology, nor the beautifully stimulating spaces. And while the learning content we use here is great, it's not at all different from the content we have access to outside DBC. The key ingredient is much more subtle. It is the frustration safety valve created by the learning ecosystem.

Much like startups are characterized by volatile emotional swings between euphoria and horror, learning difficult things is characterized by volatile emotional swings between inspiration and frustration. However, unlike startups where the key is to try to remain even keeled to maximize effective decision-making, learning productivity is much higher when the learner is inspired.

When you are learning alone and you come upon a learning obstacle, be it a poorly explained example or a misunderstanding-caused error, it might take hours to extract the intended lesson. There is a cost to getting stuck that is much greater than simply wasting a couple of hours. That cost is the consequent frustration that dampens your efforts well after you moved on.

We have internal clocks that track productivity. When we spend longer than expected on a problem, we tend to get demoralized. When we spend significantly longer than expected on a problem, we are much more likely to quit. [1]

The secret sauce of Dev Bootcamp is its frustratrion safety valve. There are over 50 students and roughly 10 staff members available to help when frustration is on the horizon. The advantage of that is twofold. First and most obvious, you are less likely to fall into a frustrated state of mind when you have the opportunity to ask someone for help as soon as you feel the need. And second, you are more likely to push yourself out of your comfort zone because you know if you fall someone will be there to catch you.

That safety net is the greatest asset DBC has to offer, and it's well worth the cost of attendance. It considerably catalyzes your learning, and that added efficiency in my opinion represents a great technological innovation.

[1] There's a great quote from Sandi Metz's Practical Object-Oriented Design in Ruby that relates to this concept of productivity-induced states of mind. If you replace the word software with pedagogy, she might as well be talking about education. "Bad software impedes our purpose and interferes with our happiness. Where once we felt productive, now we feel thwarted. Where once fast, now slow. Where once peaceful, now frustrated. This frustration occurs when it costs too much to get things done. Our internal calculators are always running, comparing total amount accomplished to overall effort expended. When the cost of doing work exceeds its value, our efforts feel wasted."

[dnlgmb]

Where had we been, we two, beloved Friend! If in the season of unperilous choice, In lieu of wandering, as we did, through vales Rich with indigenous produce, open ground Of Fancy, happy pastures ranged at will, We had been followed, hourly watched, and noosed, Each in his several melancholy walk Stringed like a poor man's heifer at its feed, Led through the lanes in forlorn servitude;

Excerpt from The Prelude by Wordsworth

[dnlgmb]

Beware What You Measure

Written on October 13th, 2012 after finishing Week 2 of Dev Bootcamp. I waited to publish it until I could see the full effect of measuring our Dev Bootcamp progress through a single metric. I'd love to read your feedback if you have any.

By the end of week 2, it was obvious. Something had drastically changed the environment at Dev Bootcamp. Our tech microcosm went from infectiously optimistic to tensely anxious, and the effects of this change reverberated through every student's learning approach. What follows is my account of this abrupt change, and my extrapolation of what it says about the entanglement of social systems and measurement.

The Source of Change

It was early on the first Friday morning at Dev Bootcamp when Shereef asked us to gather as we had now done several times over the course of the week. The purpose of this session was to poll students for feedback about the experience so far. Most important among the questions in the poll was the one gauging our level of learning comfort during the week: panic zone, panic-learn edge, learn zone, comfort zone. As a group we have agreed that the most productive learning takes place when people are at their panic-learn edge, so naturally that's where most people should have been. [1]

Everyone voted, results came in, and... the overwhelming majority of people spent their first week in the learn zone. While none of the students seemed overly worried about this, Shereef seemed pretty distraught. Like kids who are sheltered from the dangers of the world by their innocence, we are relatively blind to all the things we still have to learn to become professional web developers. Shereef knows how far we still have to go, and front and center on his mind must be that we only have ten weeks (nine at this point) to get there.

The Cause

In order to push us towards the panic-learn edge, the staff created the concept of aims. Dev Bootcamp's backbone is a curriculum of programming challenges starting at basic Ruby and branching out towards advanced Ruby, JavaScript, databases, and Rails, among other topics. The staff gave us five aim choices. The first four choices involved advancing through challenges at varying speeds and were progressively more aggressive. The fifth and last choice was to take no aim.

We were encouraged to take an aim and keep it at all costs, so most people decided to take one. And therein lies the switch that radically changed the learning environment. Up until this point, most people had been measuring their progress internally by gauging their effort level. Most would spend half their time solving challenges, and the rest of the time helping others, reading, or engaged in some other form of learning. Once aims were introduced, everyone's compass synchronized towards one direction: complete as many challenges as possible.

The Effect

The cultural change was dramatic. Most students reallocated most, if not all, of their time to solving challenges. It worked wonders in terms of accelerating most people's progress through fundamental Ruby. [2] But it had the added consequence of linking learning to number of challenges completed. And in doing so, aims introduced a measurement that turned our cartography learning model into a racetrack. Students now had instant feedback on how far 'ahead' or 'behind' they were. The problem is that number of challenges completed misrepresents progress.

Introducing aims solely focused on completing challenges had a destabilizing effect. It is always difficult to tweak complex systems because small changes can have amplified effects. If these type of aims narrow the learning horizon too much by overemphasizing challenges, or negatively impact the learning ecosystem by decreasing social interaction, they kill the goose that laid the golden egg.

Design Your Measurements

It's no secret that the act of measuring something affects the consequent decision-making of those being informed by the measurements. That's the whole point of measuring things, is it not? That is, to have a better understanding of cause and effect; to obtain the desired "effects" by adjusting the right "causes".

Measurement's ability to influence direction is its raison d'etre, but in order for measuring to be effective, you must measure the right things. If you misreprensent your priorities in the measurements you're taking, you'll end up drifting away from your inteded objective. Measure the wrong thing and you may end up veering off into dangerous waters.

Industries, companies, and teams are particularly useful subsets to study the effects of measuring. Look closely enough and you will find that what gets measured, implicitly or explicitly, has shaped the incentive structure of the individuals in the group. [3] There's a reason why your parents don't want you hanging out with certain people, even if they can't quite articulate it.

What's really interesting about this understanding of measurement's effects is the level of control it gives us over designing organizations. It isn't enough to direct people with rules or procedures. Human behavior operates at a much deeper level than the relatively simple natural language communication that takes place between people. 

No great building was ever built by an architect simply telling people what to do. Great buildings are first designed - methodically, painstakingly, scientifically, intentionally designed. Only then can they be constructed and take the form and function intended by the architect. So too organizations must be meticulously designed. Developing the intended culture depends on growing the right incentive structures. And growing the right incentive structures demands great design. Nowhere is this more true than in educational environments. [4]

How Does the Story End?

One of the things that makes the tech industry such a force of creation is its people's self-awareness. Perhaps aims were misinterpreted by me, or mis-implemented by all of us, or maybe we just needed time to internalize it into our learning approaches. Whatever the case, through the staff's successful architecting, we have managed to tweak aims to include a broader part of the learning spectrum. We should all be so lucky to be able to hypothesize, experiment, and adjust in all facets of our life. Particularly our education. 

Notes

[1] Note I am qualifying the learning that takes place at the panic-learn edge as the most productive for learning something new. By most productive I mean it's the most efficient method in terms of amount of learning per time spent. Whether this is the best possible learning approach outside of the context of a ten week, intensive course of study is, at least in my opinion, yet to be determined. I have an inkling that a playful state of mind based in your learning zone is more conducive to mental mapping and understanding. However, if you want to go from zero to web developer in only ten weeks, you can't (always) stop and smell roses.

[2] Setting goals/aims is great for supercharging progress on a specific goal but the cost is you ignore everything else. Aims have an inherent ability to alter a system's incentive structure, may be positive, may be negative. Look no further than Wall Street. What gets measured and rewarded is short-term gains and aggression - guess what the focus of the whole industry has become? Aggressive short-term profits with complete disregard for long-term stability and growth. I think most people on Wall Street intuitively know this, and that's why there's so much cynicism there. That at least was certainly my experience.

[3] Measuring can be implicit or explicit. If you are specifically tracking something, then you're getting quantitative feedback on its progress. However, most measuring that takes place in social systems is much more subtle than statiscs and metrics. The values embedded in a culture are implicit measurement mechanisms that people use, consciously or not, as steering systems.

[4] Most American schools have disastrous incentive structures. If you attended one, you have first-hand experience. If you didn't, you may be interested in an essay by Paul Graham called Why Nerds Are Unpopular.

[serialmilk]

I got to enjoy my favorite kind of learning moment today: when an innocent, seemingly simple question turns into a journey, which leads to a revelation. The revelation is that code is poetry, not prose.

I was really dissatisfied with code I wrote at the end of last week for a flashcard program...

[dnlgmb]

Notes from 2012 Startup School

These are my notes from YCombinator's 2012 startup school at Stanford. If you can't tell exactly what a note means, feel free to ask me or use your imagination.

Abstract: 

Don't start a company for the sake of starting a company. A startup should be a means to an end, and the best ends are those based in fundamental human wants/needs. Particularly those wants/needs that you have and are adamant about meeting somehow.

There are two feelings in startups: euphoria and horror. And neither of them accurately depicts the reality you're facing. Be able to deal with this emotional dichotomy (relevant line from If by Rudyard Kipling, "If you can meet with Triumph and Disaster; And treat those two impostors just the same").

The biggest killers of startups are (1) not making something people want, (2) founder disputes, and (3) giving up too soon.

Table of Contents

Mark Zuckerberg (Facebook)

David Rusenko (Weebly)

Joel Spolsky (Fog Creek, StackExchange)

Hiroshi Mikitani (Rakuten)

Ron Conway (SV Angel)

Tom Preston-Werner (Github)

Ben Horowitz (Andreessen Horowitz)

Ben Silbermann (Pinterest)

Patrick Collison (Stripe)

Jessica Livingston (YCombinator)

Travis Kalanick (Uber)

__________________________________________________________

Mark Zuckerberg (Facebook)

School emails were a way to ensure people used their real identity online (rare on the internet in 2000)

Now (present day) self sustaining culture of real identity on internet

Originally Facebook included course data; they had to parse course catalogs every time they expanded to a new college; debate with Dustin about quality of service relative to this course data

Colleges were the validation strategy for the concept of an online social network

Didn't wanna burn a lot of money so they rolled it out college by college

Confirmed: It was in fact Eduardo's job to sell ads

$85/mo server (main constraint on growth)

New features that worked (baking period)

Took 1yr to get million users (good time to bake/figure things out)

Cared more than bigger companies about making it exist

Launched Facebook in schools with online social networks to see if Facebook was better (the idea being that if Facebook is adopted in these schools, it would definitely be adopted in schools without a social network)

Paul Graham asked Mark: If he was a couple of years younger and Harvard had had a social network would he still had built Facebook? Answer: hard to know

On the subject of technological progression: Wikipedia was about putting humanity's information, and then Google about searching for that information in better ways, then Facebook about adding to that the world's social information... next thing will be even better, so there's a trend to this growth of sharing info (they stopped just short of discussing what the next big thing on the curve will be...)

He started Facebook because he wanted it and thought it should exist globally

PG said (paraphrased): I wish there were more founders who got started by their companies rather than the other way around (meaning that the founders who have an itch that they can't help but scratch are in many ways better than those who simply want to start startups as an end in itself)

On that note Mark said (paraphrased) "Explore what you wanna do before committing"

What kept people coming back to Facebook? What makes a human human, people to people connection (psychology); most humans when dreaming dream about social interaction; Google indexes info but not social info; technology extends the nature of human interaction (Steve Jobs said that computers extend your mind and along that same theme Mark believes social networks extend your social capacity)

Advice: do something that's fundamental to what it means to be human because the most interesting things are in those fundamentals

__________________________________________________________

David Rusenko (Weebly)

[article by David with a similar message as this talk: The Importance of Launching Early and Staying Alive]

He thought there should be a more simple and powerful way to create a website

Started with: why is it so hard to build a website?

6 months after starting... only 12 users

8 months after starting... only 30 users

Accepted into YC W06

Then TechCrunch wrote about them (they got TechCrunch'ed) and userbase ticked up to 300 users before settling back down to 100

11 months after starting they're still at less than 100 users

aside { Always took saturday off but every other day worked long hours }

aside { He showed a picture of an analog tracker of Weebly's userbase that read Total Users 6913 and Logged in 3 } -- the ones logged in were the three founders

April 2007: bank account = $100  (14 month after starting and user count is trending down)

Featured in Newsweek 15 months after starting and user ticked up to 400 before going back down again

Featured in Time 18 months after starting, but still not growing

...long work time with no real positive results...

...But they knew this was a good problem to be solving

Unfortunately the metrics had not yet proven them right

Finally 20 months after starting they see growth, first real traction just via word of mouth

29 months after starting they move into their first real office

Fall 2008 was a bad time to raise money, they ask themselves: how much runway do we have?

34 months after starting: bank account balance going down

They map out their bills on the calendar and prioritize them into two types: those that would shut down the service if not paid, and those that would just charge a late fee

35 months after starting: finally break even

4 years after starting: finally over 5000 users and developed some growth momentum

Now 2% of all internet websites are built on Weebly

Weebly has a net promoter score of 80% (which means costumers are extremely satisfied)

Message: this is still an unfinished story

Whats next?

Today regular people are just as frustrated, just as confused when building their websites

And people just want to showcase self, or store, or anything

Lots of work left

Advice: You can't succeed if you quit

Behind every overnight success there are years of work

__________________________________________________________

Joel Spolsky (Fog Creek, Stack Exchange)

You don't have to raise VC money, you can bootstrap

Started Fog Creek because he couldn't find a cool place to work

Back in the internet boom many there were many 'stupid' companies

His model was proprietary software and consulting (consulting would fund the proprietary software they were building)

There was huge demand for HTML builders (lots of demand to build sites)

With revenues from that they bootstrapped the business

They made $250K in 2 months

In November of 2000, billings of consulting companies went down by 90%; those companies didn't realize that the market had collapsed until April 2001

At this point Fog Creek started selling Fog Bugz, bug tracking software

They made enough money to eat

Numbers started going up slowly but consistently

At first $15K/month, then $25K/month, and finally in 2003 they moved into a legit office

They created a recruiting pipeline from interns

Now he shifts into talking about StackExchange:

The goal of StackOverflow was to kill expertexchange "because they were evil"

He got together with Jeff Atwood

aside { Took 10 years to build his blog's audience, Joel on Software }

There are 2 ways to build a business: get big fast, or organic growth [he does a great job of explaining the details of these two strategies on this blog entry]

Get Big Fast => make mistakes fast and iterate = most efficient way to discover business model

Organic Growth => mistakes can kill you

Organic is much more careful

But at the end of the day, failure to decide is what will kill you

Anything you do to defer raising money is great for leverage

aside { He only works at StackExchange now, CEO }

Fog Creek is a cash cow through Fog Bugz

Trello, Fog Creek's new web-based project management app, is an instance of Get Big Fast because it's a land grab

Land grab implies that there are strong network effects and once someone gains significant market share, they're very hard to displace

Trello is free for that reason and they can figure out how to make money later

They can use Fog Bugz cash cow to fund Trello's growth, which means once need to raise VC money they'll have full leverage

Advice: You have to commit to either grow organically or get big fast

__________________________________________________________

Hiroshi Mikitani (Rakuten)

Founded Rakuten in '97 and wanted to partner with Paul Graham's Viaweb to make an internet mall for Japan

Rakuten is top 5 in Japan for online commerce (#1), ebanking, etravel, and others

One of the main differences between Japan and Silicon Valley is that Japan has very little in terms of a VC ecosystem; there are investors but they don't act as advisors as much

Hiroshi created the company in such a way that they would need to rely on VC money to fund their growth. Their cash flow system would allow them to grow

note { it would have been nice to know how that cash flow system worked, since of course growing this way is every entrepreneurs dream }

Rakuten has been cashflow positive since the beginning

Advice: be patient when searching for the business model

To find Rakuten's he relied on the idea that there was a need for better communication between consumers and merchants

Rakuten acts as a liason with retailer

He doesnt think about competitors much because you can't control what they do; but you can control how you improve yourself, so focus on that

Rakuten's growth is a result of organic growth and buyouts

They also focus on brand building through sports sponsorships

Their acquisitions fall into two types: geographic expansion and strategic

The main concern of an acquisition is mission and culture integration

Founders and management must buy into Rakuten's culture and mission because if they don't and just leave soon after the acquisition, the integration falls apart

Rakuten recently invested in Pinterest. They see lots of potential for e-commerce in Pinterest's approach to social pinning. It's very graphical and strong influences buying behavior

He thinks Pinterest is the best social network for e-commerce

Big and exciting challenges the near future of e-commerce. This is only the early stage

Logistics and automation are two of the biggest, followed by devices and mobile tablets. Also social shopping, which hasn't been successfully executed yet

Hiroshi also mentioned that one asking questions to professionals is much more powerful in the shopping experience than asking friends

aside { He learns from competitors but doesn't fear them }

He likes hiring younger people at the early stages of a venture to reinforce culture, and then follow that up with hiring experts for esperience

__________________________________________________________

Ron Conway (Angel Investor)

Ron is part of SV Angel and one of the most successful investors in Silicon Valley. Check out his bio here.

Has funded ~650 startups since 1994

He co-founded Altos Computer in 70s, which went public in '82. Back in the 70s you needed growth and profitability to get VC money

However the things that Jessica talked about were as true in the 70s as they are today

Important to have: Determination. Conviction. Leadership

One things that's different: people drink less. Back then work hard and play hard, sometimes at the same time. Now startups segment those two better...

Back then it wasn't about customer satisfaction as much as today

Best decision he ever made was to invest only in software (since he started in 1994)

He got in at the basement of internet investment

He really enjoyed/enjoys mentoring more than being an entrepreneur himself

Wife thought he was crazy as usual

Ben Rosen said innovation is all about growth and disruption. That's how they went into internet investment

He thinks we're still in the infancy of the internet

For one, e-commerce is coming in a much bigger and impactful way

Some of his biggest misses: Salesforce in '98, Pandora, Palantir, Kickstarter

Paul asked, is there a pattern to what investors miss? What do investors not get?

Ron answered, typically the first of something

He invests in the entrepreneur first

When he invests in entrepreneur he invests lifetime because you develop a great relationship that making communication and mentorship very easy

For example, he invested in Twitter without due diligence becasue he had invested in Odeo and believed in Ev Williams

Paul asked, What'd you think about the idea of twitter? Ron responded with, never argue with metrics and growth

Don't think about monetization that much, having 200,000 users will take care of that for you

Google had no idea how they'd make money at first

They said we have awesome tech and were honest about not knowing how to monetize. Even in meetings with VCs, they wouldn't even include a spreadsheet showing revenue projections

Larry and Sergey were honest about not knowing how they were gonna monetize

Their technology was/is based on relevance and page rank, both of which were novel concepts at the time for web search

Paul asked, what did you think when you first met them? Ron asnwered:

He heard of them from a Stanford professor while they were both trying to forget they were wearing tuxedos at a christmas party. What he saw when he met them was two great scientists who were very strategic and determined

Larry and Sergey asked Ron to get Sequoia and to get Kleiner-Perkins (John Doerr specifically) because those were the main investors in Yahoo and AOL -- this shows strategic thinking

After 1 month Google was exploding and Larry and Sergey were frustrated with Sequoia and KP. Ron communicated as much to Sequoia and KP and told them Google was ready to walk away from them. This put pressure on the two VCs and they closed the deal. Ron said he knew Google funding was history when the agreement finally closed

Ron admitted that he didn't understand Facebook, but he knows growth and you can't argue with growth metrics like Facebook was showing

He told a story of the first time he met Mark. The meeting was setup by Sean Parker, president of Facebook at the time. He asked Mark if he knew he was dealing with an edgy person in Sean, and Mark said, "Yea, I know, I figured that out." Immediately after Mark got to the point of the conversion and simply asked Ron what he could do to help Facebook. Ron said he could help with funding (at this point Facebook had only taken the Peter Thiel investment). Ron helped advice the VC round

Don Graham (of the Washington Post) offered $50M and Ron told Facebook to take it

Paul asked Ron is any VC said no to Facebook, to which Ron responded with a solid Yes. He was however unwilling to mention who

Going back to Google, Paul asked if Larry and Sergey knew they'd be as famous as they now are?

To this Ron answered with a story: Napster was so disruptive that 3 magazines put Shawn Fanning on their cover - on the same month. Larry and Sergey were introduced to Shawn at a party and Ron introduced them as the founders of what will become the biggest company on earth some day. After meeting Shawn, Larry and Sergey expressed frustration to Ron because they would never be as famous as Shawn...

SV Angel invests in people with personalities, leadership, charisma, an ability to convince people to work with them, determination, fearless, communication, drive, and 50 other traits

When he invests in an entrepreneur, he wants to invest in all his/her companies

Product focus!

If product is good, customer is happy - must be a craftsman obsessed with perfect products

Advice: Make your customer happy

__________________________________________________________

Tom Preston-Werner (Github)

Tom started his talk by asking: what does it take to raise $100M? (Context: Github just raised that amount)

But then he said, the right question is not about money. It's not about how to raise money

"We're not trying to get more money"

What does matter?

A company is nothing except the decisions that it makes. Decisions are made by people

Therefore the only thing that matters is people

And those people had better be the right people

Founders should have complementary skills

At the beginning of Github people wanted to pay for it to ensure its continued existence

One of Github's first employees was a git expert whom they met at a Ruby meetup

They continued to assemble a broad diversity of people with complementary knowledge. None of them were business people. None had created large businesses. None of them were executives

note { Executives know too much. Dont think different enough. Not knowing something is powerful }

aside { Getting experience at other companies is good for what not to do }

Get together with people, have a beer, and talk through problems

When faced with significant issues, step back and evaluate it

Know that you can't effectively move forward if you don't solve those issues

Know that you'll screw up hiring. It's inevitable. Be ready to let someone go

Things are going so fast that you'll have to clean stuff up as you go

Titles are crap. There's too much to be done to specialize at the beginning

Do you have the right mix of personality types?

Tom considers himself logical and pragmatic

One of them is the product visionary

Another leans more towards business and operations

Yet another leans more towards culture and happiness

When hiring, the crucial question to ask is how is this new person going to help push the company forward

People are the only thing that matters

But wait, actually product is the only thing that matters

Start with design. It's how your customers interact with your products

Go for a user experience that feels natural, like the design of a car (you don't really think about driving when you drive. Pushing on the gas pedal to accelerate, moving the wheel to steer, etc. It all feels natural)

What you exclude is just as important as what you include

Focus over features

Know that you can't do everything. Those things you do should be great

One founder must be design minded

To increase focus, have a company mission that can be expressed in one sentence

First Github mission: "Git hosting. No longer a pain in the ass"

Then more about collaboration: "Make developers' lives better every day"

Then: "We use Github to build Github"

Now: "Making it easier to work together than to work alone"

And that's what they're gonna do with the $100M...

There's still a lot more room to fix collaboration, even outside of software

If people like your product and you have lots of users, you have to make the decision: do you preserve a stable niche business or do you go for world changing

When you face that opportunity, don't squander it ("like the show Lost did"... zing!)

For Github it is the opportunity to reinvest all their profits back into the company to make Github even better and more impactful

Almost all products out in the market are crap

But wait... Philosophy is the only thing that matters

People they hired created the culture

Reinforcing the core values: optimizing for happiness; best argument wins; not about ego; keep an open mind; avoid politics

Work from first principles. Redo things from scratch if you have to

Dont just copy things, ideas, or advice. Think about them in the context of your company

Create super fans out of your users

Github at some point measured the success of their customer support by counting exclamation marks in customer's help emails

Be awesome and change the world

Adivce: a strong culture can be used to hire

Which one is only thing that matters? Is it people? Is it product? Is it philosohpy?

This just just a false trichotomy...

It's like three quarks of an atom. They cannot exist on their own

Master these three and you will have the answer to the question, how do I raise $100M?

__________________________________________________________

Ben Horowitz (Andreessen Horowitz)

Ben started off with some Buffet wisdom, which means advice he got from the man himself: The greatest thing about being rich is that you don't have to do business with people you don't like

He followed that up with some Michael Jackson wisdom: It's harder than it looks (referring to Ben's question about a certain dance move MJ performed during The Wiz. Specifically when he spins backwards while going up the yellow brick road)

aside { Context: he went over the grim statistics of startup success. Then said, they got the drop out trying to keep the kids in school }

Technology means better ways of doing things. An innovation must be 10x better because a little better doesn't induce change.

Two things: You must (1) build a great product and (2) take the market

If you don't take the market, someone does and now you have to be 10x better than them

Then he went into the story of his company Opsware (theme here: startups are a roller coaster)

They had only 4 weeks of cash left just before they IPO'd

And that's after having raised an $800M Series C round of funding

So they IPO at $400M 18 months after starting

Then the first quarter after the IPO he has to reset guidance down to 1/3 expected revenue

They were burning $25M per quarter and had to go through several layoffs

He slept like a baby: woke up every 2 hours and cried

Then he thought (to try and make himself feel better): how bad could this really be?

Bankruptcy, layoffs, disappointed investors, customers screwed -- very bad, that didn't help

In startups you experience two emotions: euphoria and horror

Next part of the talk: Is it too late? Is innovation dead? Is venture capital over?

He shows a quote from the year 1843 by Henry Ellsworth: "The advancement of the arts from year to year taxes our credulity, and seems to presage the arrival of that period when human improvement must end."

Another quote, this one from 1936 by John Maynard Keynes saying that once man's basic needs are met, people won't wanna work very hard and will only spend on needs leading to lots of savings

Ben says that the line between wants and needs doesn't exist

Tech increases needs

Wants become needs

And there's no important disctintion between needs and wants

Then he referenced Nicholas Carr's claim that IT had become a commodity in 2003 and asks, "how could someone from Harvard say something so stupid?" Then says, "only someone from Harvard could say something that stupid" (consider this is before social networking, etc)

Next part of the talk: How to build a company and what does Andreessen Horowitz look for 

Tech companies folow an extreme power law curve

15 of all startups started in a year will ever get to $100M in revenue

And those 15 will be responsible for 97% of all market capitalization of venture capital

Then asks, "Is your company one of those 15?"

What does it take? First, a breakthrough idea (which is very hard). These breakthrough ideas look stupid at first, which is why no one else is building them (so don't be discouraged if you tell someone your idea and they think you're nuts)

The line between vision and hallucination is very fine indeed (and breakthroughs come from hallucinations)

All great investments look stupid at the time

They also look for a founder with the courage and skill to build the company

This starts with the courage to have the idea in the first place. It's the same kind of courage it takes to build a company

The courage needed to push an idea that no one believes in is the same needed to get the company through the hard times

Skill = Can you build a product, and over time gain the skills to run the company?

Leadership vs Management

Almost no founder has management skills at first

But the ones who succeed have leadership qualities: can you articulate a vision that people wanna follow? Can you convince your employees you have them in mind? Can you articulate a vision and get people to follow you? Do you have the courage to get through the hard times?

And lastly, don't get involved with the purple drank

__________________________________________________________

Ben Silbermann (Pinterest)

It can take a long time to build awesome things

Running a startup is not really a marathon because in a marathon you know where the finish line is. In a startup you have no idea when the roller coaster will end. They're about full uncertainty

Ben worked in Google's customer support until 2008

His whole family works in medicine, but after college he decided he didn't want to be a doctor. He kept going back to the idea of building a product

Committing matters

That means that building side-projects won't be enough

One of the deciding factors in Ben finally committing was his girlfriend's ultimatum: either do it or stop talking about it

The first product he tried was an app for iphone catalogs (called Tote)

Advice: there are lots of ways for investors to say no

3 lessons if you need funding: (1) even rich people are subject to free cookies (their attention span is that of a normal person); (2) investors are just people, so they might be wrong; (3) if you really need the money, and they know you have no leverage, you'll get taken. Give them reasons why you have the leverage. This thing you're building is big, and others want the deal

People will give you all kinds of advice, but tech is a high volatility industry. Fundamentally the future is unwritten. They don't know for sure, so assess for yourself whether their advice applies to you. They might be wrong.

Be great at one thing

When to ship: ship when you have one thing you're proud of. It must be worth your user's and your own time

Pinterest collection must look cool for people to want to share it (sharing reflects on a user, so if it isn't cool, they won't share)

At the beginning, they actively pushed Pinterest for 4 months and it resulted in only a few users (although those few really loved it)

Then they started focusing on marketing by meetups because Pinterest is about findig people who share your interests

Then growth started

aside { Investors have access to the same information we do; no magic; no secret sources... for the most part }

He thought the only way to do this was to hire super genius people. But there are lots of ways to succeed

There are lots of different kinds of people in the world

Just trust the data, your users, and your own instinct

His take on Pinterest: it's a tool to find your inspiration

Adaptability is fundamenfal

Pinterest is a team of talented and diverse people

Ben says: although we tend to have an idea of inventors as lonely people, and we also tend to have a negative idea of design by committe, some of the best things in the world are made by groups of people

Advice: (1) build something you believe in because otherwise you'll burn out, (2) don't give up. Don't let someone talk you out of your dream

Silicon valley is a weird place. Everyone talks about doing startups. But starting is lonely and it's easy to think you need to work harder. NOT TRUE. Be social and live. That's how you get through the tough days.

__________________________________________________________

Patrick Collison (Stripe)

Startups are not rockets, it's a lot more about the late night talks, arguments, details

How growth works:

Oct 2009

Patrick and his brother, John, were fascinated by online payments and the fact that the existing companies tackling online payments were dinosaurs with crappy services

Stripe was actually called /dev/payments at first

Everyone else in online payments was targeting businesses. They thought people should be the focus of the user experience

So they took off to Buenos Aires, Argentina (because it was winter in Boston) and started coding

Strip is an unusual startup in that success is about tech but also about payments and finance

And the had no experience with the finance stuff

For the first 6 months transaction volume was nonexistent

Then Peter thiel invested

...and transaction volume was still really low through the first year

The beta period is like a baby blanket, very important for figuring things out and get a good foundation under the company's feet

Advice: make 100 people very happy

Their original customer service support was one on one. They set up a chat room on the website to ask questions. If someone's question was not answered in 30 seconds they would get a notice and one of them would log in to answer it personally

They didn't want to build a thin software layer, they wanted a full experience

aside { The unglamorous side of startup is the sleepless nights, the stress, all the doubts }

Greg LeMond quote: It doesn't get any easier, you just get faster

Two and a half years later, Stripe finally has lots of traffic volume

They are now a 34 people team

Analogy: shipping physical goods in containers was a huge innovation. Before containers transportation for physical goods was a huge issue and very expensive. 10% of the cost of all goods was transportation alone. Now containers have cut transportation costs by 95%

=> The elimination of friction facilitated the creation of manufacturing hubs and reshaped the world economy

It's the same with the internet's economic infrastructure

The internet has revolutionized many things, but what about the way we transact?

Stripe wants payments to be a ubiquitous utility

But keep in mind, most of your startup time will not be spent dealing with these kind of fundamental, high-level problems, but these broad, all-encompassing ideals and vision is what keeps you pushing forward

All the high level ideas plus the details and implementation are both really addicting and that's why you keep doing this

__________________________________________________________

Jessica Livingston (YCombinator)

aside { this was my favorite talk of the conference }

YC has funded over 400 startups since starting in 2005

Jessica showed a diagram with lots of people (representing startups) on the left side, a funnel in the middle, and a few successful company logos on the right. There were a lot fewer logos than there were people. The theme of her talk was, what happens inside that funnel that kills so many startups?

#1 Weapon = { Determination => Resilience and Drive }

There will be lots of rejection, expect it. And everyone will have doubts about your idea, about your ability to execute, about anything that can be doubted

We are used to external validation and to some extent we crave it. From the time we go to school to when we get a corporate job, people are reinforcing the well-worn path over the path less traveled. It's hard to swim against the current when we have only trained to swim with it.

And no one is impressed by the ugly duckling, which is essentially what every startup is before growing into a beautiful swan of a company

For example, Airbnb was on the brink when they arrived at YC in W09

Keep in mind that all great ideas seem crazy at first, and good execution will make people's eyes open

Even YC got rejected when they first got started (she showed some text messages from Ron Conway that showed him blowing them off a bit)

Again, if you execute well you'll eventually convince people of your vision

There's no playbook, just improvisation

Cofounder Disputes are the second biggest killer of startups

You must know each other and deal with any red flags before they blow up

Advice: Investors have a herd mentality; none will want to fund you until one wants to fund you, in which case they'll all want to fund you (kind of like chicken and egg problem)

You have to create a competitive situation when fundraising

It's not a deal until the money is in the bank (Jessica told a story of a pair of founders who sold everything, moved from Houston to Silicon Valley, only to have funding pulled by the VC who got them to move once they arrived in SV)

She said, "Fundraising is a bitch". Then apologized for swearing. Great moment.

Beware of distraction. Corollary: stay focused

3 things to focus on: (1) write code, (2) talk to users, and (3) exercise

Don't talk to corporate development teams. It'll feel flattering that a big company is interested in you, but they're probably just looking to make an HR acquisition, which is lame. Especially once you spend countless hours talking to them and they low-ball you with a crappy offer. Having spent so much time on this will deflate your ambition and may cause you to give in prematurely.

Number 1 killer of startups: NOT making something people want

SO MAKE SOMETHING PEOPLE WANT

Starting a company is a roller coaster, and there is nothing to dampen you from experience every single bump on this unpaved road

She talked a lot about the lows, but also mentioned one high example: Codeacademy pivoted at the end of their YC batch and grew to 200,000 users in 3 days 

No extreme ever lasts, and people watch you on the roller coaster, so you need to have thick skin

__________________________________________________________

Travis Kalanick (Uber)

Uber doesn't own any cars nor do they employee any drivers

Their growth, revenue, and satisfaction metrics are phenomenal

Uber is showing exponential revenue growth

At launch people didn't believe in the idea

Liquidity?

Two keys to Uber's success: process management and product management

Operations are key

Quality and choice (low cost option for lower activation energy and deeper engagement)

They have a team of mathematicians, physicists, etc. to predict demand in real-time. Their app even allows them to know when a user opens the app to view what's available

How do they make the magic happen? Math! => demand prediction and dynamic pricing that lead to market clearance at any given time

Drivers have a heat map of demand prediction generated with supply considerations in mind form maps for driver guidance

He showed some data on how events (like a Giants game) affect demand. In fact, in San Francisco demand increases when the Giants win, whereas in Boston demand increases when the Red Sox lose 

User experience: the drivers give riders high fives when they get in (this reminds me of my guiding principle of PIEf = promote inspiration, eradicate frustration)

User experience: They gave their drivers flowers to hand to women riders last Valentine's Day (emotional connection)

User experience (creating fans): Last President's Day, 1 out of 20 riders was picked up by an Ubercade, which is like a presidential motorcade (secret service and all)

Some users even give drivers hugs... they love the service

Travis also commented on the regulation landscape and its bureaucracy... "It's corrupt out there". To be disruptive you have to be willing to fight, so don't be shy

[flavorite-blog]

18 days into DevBootcamp, I realized today that eventhough I’ve had conversations with (almost) everyone in the cohort, and I don’t know what most of them were doing before DevBootcamp. Startups? Finance? College? I don’t know. Marketing? Sciences? Programming? I don’t know either.

And this is...

[dnlgmb]

You never change things by fighting the existing reality. To change something, build a new model that makes the existing reality obsolete.

Buckminster Fuller (written on one of the columns at DBC HQ)