<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Sam Gerstenzang on Medium]]></title>
        <description><![CDATA[Stories by Sam Gerstenzang on Medium]]></description>
        <link>https://medium.com/@gerstenzang?source=rss-996a11626697------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*VfWdiDXap-lhg4RsNemO6Q.jpeg</url>
            <title>Stories by Sam Gerstenzang on Medium</title>
            <link>https://medium.com/@gerstenzang?source=rss-996a11626697------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Mon, 16 Nov 2026 04:40:53 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@gerstenzang/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Lessons to my younger self (and other young people today)]]></title>
            <link>https://medium.com/@gerstenzang/lessons-to-my-younger-self-and-other-young-people-today-c245ed938391?source=rss-996a11626697------2</link>
            <guid isPermaLink="false">https://medium.com/p/c245ed938391</guid>
            <category><![CDATA[life-lessons]]></category>
            <dc:creator><![CDATA[Sam Gerstenzang]]></dc:creator>
            <pubDate>Tue, 01 Jan 2019 19:34:49 GMT</pubDate>
            <atom:updated>2019-01-01T20:47:49.503Z</atom:updated>
            <content:encoded><![CDATA[<p>One of the challenges of being young is you don’t know what excellence looks like — you just don’t have the reps to know. So you should make it a priority to shortcut those reps in any way you can— read the books, find the people, look at the work. Over and over again.</p><p>Learning from failure is overvalued. Understanding success and more importantly, why it was successful, will create a useful mental model to jump off from.</p><p>But equally important to understanding excellence is not to be paralyzed by it. You have to just <em>do</em> — while being constantly aware of the gap between excellence and your work.</p><p>This is enormously difficult because fear gets in the way. Will I ever be able to achieve excellence? Am I enough? Self-sabotage sets in at this stage — if you never try you’ll never have to witness the valley of inferiority.</p><p>You must overcome by separating your ego from your work– your work is not you.</p><p>Most young people fail because they 1) don’t understand what excellence looks like, 2) don’t do the work (due to fear, laziness, or disinterest) or 3) don’t have the skill. You can only control parts 1 and 2, and if you solve those, you may solve part 3.</p><p>Above all — do the work. Even– especially — outside of work. You don’t know enough to do the job well just during work hours. And do the work that you say you’re going to do — that alone will differentiate you.</p><p>But mastering skills is only one part of your evolution. “Experience” has three components:</p><ol><li><strong>Ability to predict future outcomes, usually based on past data</strong></li></ol><p>To predict future outcomes better, you need data (again — read, listen, watch– and find shortcuts to do so) and to constantly try to understand <em>why.</em></p><p>Predicting outcomes allow you to make better decisions today. Be mindful to pay attention to whether the decision at the time was the right one — ignore the result itself — only look at the decision process.</p><p>One of the advantages of being young is you have not yet learned lessons that are wrong. You are less likely to be an average of your experience, and much more likely to spike in ways that others won’t. This is a huge advantage of the young — and those who age well are often good as good at forgetting earlier lessons as they are at applying them.</p><p>2. <strong>Behavioral — do you know how to act? How to run a meeting properly, how to make sure deadlines are met, etc</strong>.</p><p>To learn good behavior, models are once again, helpful. Learning good behavior is best learnt from other people — find good people to spend time with — in and outside of work. In the absence of direct experience, I’d suggest watching <a href="https://en.wikipedia.org/wiki/The_Defiant_Ones_(documentary)">The Defiant Ones</a> as a place to start.</p><p>Confidence is another strange thing in young people — I see young people way too confident and way too unconfident. Confidence is like compensation — some people have too much, some people have too little — and no one has precisely what they deserve. I’m not sure about you in particular — but the right mix probably looks something like a belief you have to paddle hard to be in the room you deserve to be in.</p><p>Find people who will answer your questions graciously, and pay them back with loyalty. (Thank you.)</p><p><strong>3. Emotional control — how do you act, really?</strong></p><p>To learn good emotional control will take your lifetime, and everyone will require a different journey. My personal toolkit is as follows:</p><ol><li>Read <a href="https://www.amazon.com/Wise-Heart-Universal-Teachings-Psychology/dp/0553382330/ref=sr_1_1?ie=UTF8&amp;qid=1546010748&amp;sr=8-1&amp;keywords=jack+kornfield">The Wise Heart</a> by Jack Kornfield, and <a href="https://www.amazon.com/Power-Now-Guide-Spiritual-Enlightenment/dp/1577314808/ref=sr_1_3?ie=UTF8&amp;qid=1546365570&amp;sr=8-3&amp;keywords=the+power+of+now">The Power of Now</a> by Eckhart Tolle.</li><li>Log every extreme emotion you have, as close to the occurrence as possible. Use this log for step 3.</li><li>Find a coach or therapist you jell with. It’s a highly leveraged way to accelerate your emotional development.</li></ol><p><strong>How do you seek advice?</strong></p><p>One mistake I’ve seen when younger people ask me for advice is that they ask either a) things I can’t possibly answer or b) questions already well answered. For example, “I’m a generic smart person but without deep skills. What should I do?”</p><p>That’s a difficult question to answer. I’d suggest you seek to ask a different question: “Here’s what I know. What am I missing?”</p><p>This requires more preparation on your part — to understand what you know — but provides a more fruitful answer. Most advice of the variety “here’s what you should do” is not very useful. But providing a window in to what is unseen can shift your world.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c245ed938391" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What we’re building at Umbrella]]></title>
            <link>https://medium.com/@gerstenzang/what-were-building-at-umbrella-13ea31a6b540?source=rss-996a11626697------2</link>
            <guid isPermaLink="false">https://medium.com/p/13ea31a6b540</guid>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[aging]]></category>
            <dc:creator><![CDATA[Sam Gerstenzang]]></dc:creator>
            <pubDate>Sun, 02 Sep 2018 23:47:34 GMT</pubDate>
            <atom:updated>2019-08-23T15:51:26.881Z</atom:updated>
            <content:encoded><![CDATA[<p>We want to tell you about three trends that are changing America today.</p><p>The first is that our country is getting older. You’ll hear things like “10,000 people turn 65 a day” but that doesn’t adequately convey the actuarial drumbeat. There will be <a href="https://www.citylab.com/equity/2017/01/can-americas-aging-population-stay-in-their-homes/512009/">80 million Americans 65 or older by 2035</a>.</p><p>The second trend is that<a href="http://www.nbcnews.com/id/45376299/ns/health-aging/t/aging-place-most-seniors-want-stay-put/#.W4wwnZNKjOQ"> most of these older adults will want to stay in their homes</a>. This is called “aging in place” — the desire to live in the house and community you’ve always lived in— not an isolated, age segregated facility.</p><p>The third is that <a href="https://healthtechmagazine.net/article/2017/06/seniors-begin-embrace-digital-age">older adults are becoming more digital</a>. But most products aren’t built for older adults. Every demographic has unique digital needs — your baby sister uses Snapchat differently than your husband— but older adults have been a neglected digital consumer.</p><p>Those are indisputable demographic truths that Umbrella is building for.</p><p>Here are two things that we believe that others don’t:</p><ul><li>Products for older adults should be built and sold directly to older adults. We’re building one of the very few “direct to older adult” (D2OA) brands. We don’t sell infantilizing products to adult children buying for their parents — we sell to our users.</li><li>There is a deep power in communities supporting themselves, and technology can be used to elevate the best of human kind — not replace it. You won’t see any humanoid robots from us.</li></ul><p>We believe in a world where technology can allow all of us to live independently, supported by our community, in comfort and safety.</p><p>Our first product is a ‘do-good’ marketplace that allows the community to pitch in and earn extra income.</p><p>We do all kinds of tasks in and around the home to make it easier to stay there — everything from raking leafs, to snow blowing, small household repairs, house cleaning, technology tutoring, and transportation to appointments.</p><p>The side effect of serving our customer is the creation of purposeful work for our communities. More than half of the people who work through Umbrella are themselves retired but want to stay busy and put their skills to work. Our Members and our Neighbors (what we call the the folks who do the work) form long lasting relationships through the process– sharing lemonade, lunch, and live music.</p><p>We think this is tremendously exciting, but it’s only the first step for the Umbrella.</p><p>If you think about the world from the point of view of the modern independent older adult — who is increasingly digital, and whose needs are changing and intensifying — we can flip the idea of the centralized, dehumanizing older adult facility on its head, to create a better life for the future ahead of us all.</p><p>Lindsay Ullman, co-founder</p><p>Sam Gerstenzang, co-founder</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=13ea31a6b540" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Big E versus Little e entrepreneurship: the death of the hot dog stand]]></title>
            <link>https://medium.com/@gerstenzang/big-e-versus-little-e-entrepreneurship-the-death-of-the-hot-dog-stand-11fa36f78ff3?source=rss-996a11626697------2</link>
            <guid isPermaLink="false">https://medium.com/p/11fa36f78ff3</guid>
            <category><![CDATA[startup]]></category>
            <dc:creator><![CDATA[Sam Gerstenzang]]></dc:creator>
            <pubDate>Wed, 21 Mar 2018 02:43:16 GMT</pubDate>
            <atom:updated>2018-03-21T02:49:08.040Z</atom:updated>
            <content:encoded><![CDATA[<p>In geographical space, you can open a hot dog stand with a relatively undifferentiated product, at least compared to other hot dog stands, and have a geographic monopoly.</p><p>But on the internet, that hot dog stand isn’t going to stand a chance. On one end, you have Amazon selling very normal hot dogs, at a very good price. They’re what most people are going to buy.</p><p>And then you have the really great organic hot dogs, from that speciality online store in Portland. They sell the best hot dogs, they have the best authentic hot dog brand; you’re not going to make better hot dogs than them.</p><p>Geographic monopoly used to be a way for local entrepreneurs (little e) to create something of their own, protected by distance’s friction. You could set the rules, and your customers were the ones who decided whether you were right. This wasn’t always great for the customer; those hot dogs might have to be very bad for you to walk a bit further.</p><p>But in a more networked world, little e entrepreneurship dwindles. Geography doesn’t create a moat to protect undifferentiated businesses, you are stuck in the difficult spot between Portland and Amazon.</p><p>So instead, little e entrepreneurs drive for Uber. Big E entrepreneurs creates Uber, you drive for Uber. Amazon creates a marketplace, and you sell within the marketplace. The Big E entrepreneur creates platforms with rules, and you play within these rules to win the platform’s game. Little e entrepreneur’s customer isn’t the ‘customer’, it’s the platform.</p><p>This is great, at first, for the consumer. Uber competes against Lyft in a way that little e entrepreneurs never did or could. The question is what happens next, for you, as a consumer, but also for those would-be little e entrepreneurs who are driving for Uber instead.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=11fa36f78ff3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[17 lessons I learned in 2017]]></title>
            <link>https://medium.com/@gerstenzang/17-lessons-i-learned-in-2017-55fa5ab40e4e?source=rss-996a11626697------2</link>
            <guid isPermaLink="false">https://medium.com/p/55fa5ab40e4e</guid>
            <category><![CDATA[life-lessons]]></category>
            <dc:creator><![CDATA[Sam Gerstenzang]]></dc:creator>
            <pubDate>Sun, 31 Dec 2017 18:15:58 GMT</pubDate>
            <atom:updated>2017-12-31T20:25:33.941Z</atom:updated>
            <content:encoded><![CDATA[<p>2017 ended up being a year of tremendous growth, after what I now realize was a year of terrifying stagnation.</p><p>In 2017, I started a new company, moved to a new apartment, fell in love, started seeing a coach regularly (highly recommended), experienced loss, began developing a practice of presence, and more.</p><p>In creating resolutions for the new year, I revisited 2017’s resolutions which I hadn’t looked at since I constructed them. Some of them I had surpassed effortlessly, others were completely untouched. But the most important things I’ve done or learned in this last year weren’t on the list.</p><p><strong>On ego and work</strong></p><ol><li>Separate your identity from your work on macro level. When you start a company (or do anything that feels important), it can feel like it will become you. This was a difficult barrier to get over: is this <em>the</em> thing that can define my life? But that’s not how it works. This next project might take one year, five, or twenty, but it is not you. You are a superset of what you do, not a subset.</li><li>Separate your identity from your work on a micro level. You are not your ideas; ideas come through us and our job is to let them go. If you overlap your identity with your ideas, you risk feeling they are each precious, and of feeling criticized when your ideas are wrong or said to be wrong. If you can separate yourself from your ideas, the work will be better.</li><li>I realized my highest calling was not making the best work but producing the best work. Sometimes I’m the most talented individual contributor for a certain part of a certain kind of project. But most often I’m not. In both cases, my job is the same: how do I use the other talents inside and out of the room to create the best output?</li><li>Ultimately, we will not find happiness from the output of our work (which is static and fleeting), only in its process (which is continuous).</li></ol><p><strong>On work, growth, and victory</strong></p><ol><li>One way to win, or to make losing sufferable, is to change the time scale of victory. When others are thinking in weeks or months, think in years and decades. When others are thinking in years, use days and hours.</li><li>The real work is in the day-to-day details. That’s what consumes your life. Make sure you enjoy it. Too often we are interested in the <em>arc</em> of things rather than their <em>pillars.</em>The most fulfilling job will have both an arc (e.g. human psychology) and the pillars (e.g. the daily work of clinical care, research, etc.) that consume you.</li><li>We tend to define ourselves in personal narratives.<em> I’m terrible at parties. I’m always late.</em> Two ideas: 1) Create narratives for good. Allow yourself to tell the story of how you changed to become more like the you that you would like to be. 2) Narratives based in past behavior are by definition out of date. Sometimes allow yourself to just be open to possibility of the next moment being different. How does the story change in this moment?</li><li>Luck can’t be controlled, but your exposure to allow luck to strike can be. Work hard to increase your surface area.</li><li>The key to talent development of yourself and others: propulsion off cliffs with parachutes one size too small. Sometimes you land hard but your wings build muscle.</li><li>Previously, one of the skills I’d valued was finding arbitrages created by others’ bias. Bias against the unproven, against the old, against the boring: there are great discounts to be found in hires, investments, and in life. But one corollary I learned this year: the size of the discount doesn’t matter if you’re chasing a small opportunity. I’d much rather own 20% of something at 10% discount going to 10,000 than 20% of something at 50% discount going to 100.</li></ol><p><strong>On being present, emotions, and thinking</strong></p><ol><li>We don’t see people as they are, we see them as we are. This sounds simple but in it we find the root of all human conflict.</li><li>Meaning is created only within ourselves. This should be both terrifying and greatly freeing. Find your personal meaning and live your life for it.</li><li>Only the present exists. The past and future are fictional; they exist only in our minds. (I highly recommend <a href="https://www.amazon.com/Power-Now-Guide-Spiritual-Enlightenment/dp/1577314808">The Power of Now</a> for more.)</li><li>It is not the mind’s job to deal with feelings. You don’t need to rationalize them. Feelings are for all the information we cannot process. In general, we overrate thinking as a positive human attribute.</li><li>At the same time, thoughts do lead to feelings. Developing the skill of annotating your feelings as they occur is a powerful tool that will make you both happier and more in control.</li><li>You can use the irritants and suffering in your life to become your best self. Think of these as great gifts with which you can use to understand yourself, the sand to create your personal pearl.</li><li>The good things all have trade-offs. It is better to think in trade-offs than in dualities: there is not a best lens. How do you maximize the good parts while cushioning the bad parts of one particular lens? And then what do you get from another lens altogether? There is rarely one lens that will get you where you need to go, but each lens will bring you to a different place.</li></ol><p>Thank you for being a part of my year — I learned a lot from you.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=55fa5ab40e4e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why we didn’t start a software company]]></title>
            <link>https://medium.com/@gerstenzang/why-we-didnt-start-a-software-company-96872168eb17?source=rss-996a11626697------2</link>
            <guid isPermaLink="false">https://medium.com/p/96872168eb17</guid>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[startup]]></category>
            <dc:creator><![CDATA[Sam Gerstenzang]]></dc:creator>
            <pubDate>Mon, 23 Oct 2017 14:10:38 GMT</pubDate>
            <atom:updated>2017-10-23T14:10:58.780Z</atom:updated>
            <content:encoded><![CDATA[<p>There was a time when hardware was the business, and software was the thing that came free. The idea that you could charge money for bits instead of atoms, and <a href="https://en.wikipedia.org/wiki/Open_Letter_to_Hobbyists">build a big business</a> was crazy. Then came Microsoft, Visicorp, Lotus. And the hardware became mostly commoditized.</p><p>As a business, what made software great was its zero marginal reproduction cost. You make it once, and the cost is roughly the same if you sell it once or a hundred times.</p><p><em>Networked </em>software was an even more interesting business because it could 1) grow really fast on top of an existing network and 2) have network effects, making it hard to replicate. Low cost to provide, fast growing, hard to replicate: that’s a great business. If you find one of those, take it.</p><p>The thing is, I think we’ve reached the end of the low hanging fruit of this era. The (now) obvious networks have been built out; those sitting on top of them will wring out the last dollars. You can get paid a lot to twiddle some dials at the last great software companies, while the ad costs go up for the companies competing to place their tracking ad next to your thumbs and eyeballs.</p><p>Those companies are willing to compete against others for those ads because their own products are mostly commodities. Take one example: SMB SaaS. The expensive part of building a SaaS business isn’t building the software but selling it. You’re bidding on keywords with dozens of other companies selling nearly identical products. Stripe lists <a href="https://stripe.com/works-with/categories/recurring-payments">29 different subscription management products</a> that are compatible with its product, not to mention Stripe has decent basic subscription management built-in.</p><p>So, what’s a talented software engineer interested in changing the world to do? It’s to go back to why I, and many other programmers, began programming: to solve problems for people. I still remember the time when I had more than one concurrent user — none of whom I knew — using something I used. It felt like magic.</p><p>The most impactful thing a software engineer can today isn’t to join a software company. It’s to join a company that uses software where software has not yet reached.</p><p>Who are the users who haven’t yet been touched by software? Who are the groups who don’t have the tools or access? This is where software can make the biggest difference — and where you can make the biggest difference.</p><p>Now, software is like electricity. The most interesting companies won’t sell electricity, they’ll assume it.</p><p>So. We’re creating a new kind of purposeful work for an important underserved market. We’re hiring a software engineering lead to be our first full-time engineering hire, based in NY. I’m ready to re-hang my coding cleats. Will it be you?</p><p>Get in touch: samgerstenzang@gmail.com</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=96872168eb17" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Founder Friendly? No thanks]]></title>
            <link>https://medium.com/@gerstenzang/fuck-founder-friendly-a7dea74e0514?source=rss-996a11626697------2</link>
            <guid isPermaLink="false">https://medium.com/p/a7dea74e0514</guid>
            <category><![CDATA[startup]]></category>
            <dc:creator><![CDATA[Sam Gerstenzang]]></dc:creator>
            <pubDate>Sat, 12 Aug 2017 13:28:03 GMT</pubDate>
            <atom:updated>2017-08-13T12:53:02.965Z</atom:updated>
            <content:encoded><![CDATA[<p>The trend within the venture community over the last decade or so has been the notion of “founder friendly” — supporting your founders through thick and thin.</p><p>It arose out of a FOMF (Fear Of Missing Facebook): when Facebook’s Mark Zuckerberg passed over Sequoia because of Sequoia’s relationship with Zuck consiegliere Sean Parker and his previous company Plaxo. Who wants to make some hard decisions in a portfolio company going sideways when that founder might bad mouth you to the next Zuckerberg?</p><p>To be clear, there has been a history of experienced venture capitalists completely screwing over inexperienced entrepreneurs (I have no idea what happened at Plaxo). This is wrong, and comes from the zero sum world of private equity. This behavior should not be tolerated. But founder friendly has evolved one step further: it says we’ll support the founder through anything.</p><p>Well, fuck that.</p><p>We’re here to build companies and products. It is, and this may shock you, not about the founder. Just like when I’m recruiting my team, I want board members who will push me and the company forward. We’re here to build companies, not egos.</p><p>Sometimes founders aren’t the right people to scale the business.<a href="https://medium.com/@gerstenzang/when-to-hand-over-the-reins-b532bbc39c16"> The right CEO to go from napkin sketch to product-market fit might not be the right leader to think through defining internal incentive structures, or how to scale the creation of new product lines.</a> And I think we’re sacrificing a lot of great $10 billion, $20 billion, and $30 billion businesses because founders and venture capitalists are so stuck on the idea that it requires a founder to make an Apple, Facebook or Microsoft.</p><p>And sometimes, the founder is a massively skilled operator but morally bankrupt. Or something in the middle.</p><p>Now, this is not to say the founder CEO should be pushed around by their board of venture capitalists. For one thing, the board simply don’t have the information necessary most of the time. Operating puts you so much closer to the details, and the founder will have done so many more reps to understand their market and their company, that this intution is extraordinarily valuable.</p><p>But experiened board members have other advantages: they should know what excellence looks like across roles (even roles you’ve never met), they will have seen companies hit your phase of growth before, they will be able to see the meta-company patterns. But a board member is not a babysitter. Their role is not to be your patient friend. The job is to push you forward, much like that is your job with your own team.</p><p>Some investors take a different and important role: consigliere to help a founder get the best out of themselves. And they eschews board seats. Why? To borrow from someone else: “<a href="https://hackernoon.com/i-can-grade-you-or-i-can-teach-you-but-i-cant-do-both-505a7272e4f0?gi=18a30ca97bad">I can grade you, or I can teach you, but I can’t do both.</a>” We need both.</p><p>But when you take a board seat, and you tell me you’re founder friendly, I’m going to ask you get off my board. Because I want partner in building the company. A partner who will put in the time and effort and moral backbone.</p><p>I’ve seen too many companies because of the ego of their founder and cowardice of their board, fall far short of their potential. And this bad for everyone: the founders, the board, the employees, and their customers.</p><p>So kudos Benchmark for standing up. I just wish you had done it sooner; Uber would have been a better, longer-lasting company for it.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a7dea74e0514" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Perfect Storm: the next-generation weather company]]></title>
            <link>https://medium.com/@gerstenzang/perfect-storm-the-next-generation-weather-company-6069c45c9c21?source=rss-996a11626697------2</link>
            <guid isPermaLink="false">https://medium.com/p/6069c45c9c21</guid>
            <category><![CDATA[weather]]></category>
            <category><![CDATA[startup]]></category>
            <dc:creator><![CDATA[Sam Gerstenzang]]></dc:creator>
            <pubDate>Wed, 12 Jul 2017 12:23:37 GMT</pubDate>
            <atom:updated>2017-07-12T12:23:37.807Z</atom:updated>
            <content:encoded><![CDATA[<p>I have a suspicion that there’s a new enterprise weather company to be built. To be completely honest, I don’t know very much about weather. I started reading and asking questions a few months ago. But <a href="https://medium.com/@gerstenzang/govco-6f4eb4ad18c9">last time I tried this</a>, I met some great friends and great companies, including Eric at <a href="https://www.bayesimpact.org/">Bayes Impact</a>, Sha and Kalvin at <a href="http://navahq.com/">Nava</a>, and <a href="https://www.linkedin.com/in/akouts/">Alex</a>. So please, tell me how I’m wrong (or right).</p><p><em>Here’s the idea</em>: there have been three tremendous changes in computational technology over the last decade that have yet to be fully utilized in weather forecasting technology. Incorporating these shifts will result in significantly improved weather predictions, just as they have already begun to individually transform other fields. But here we have, dare I say it, the perfect storm:</p><ol><li><strong>Cloud computing.</strong> Weather predictions today are computed on expensive dedicated mainframes. But this computing paradigm doesn’t match the computational needs of weather prediction. The computational power needed to accurately predict weather is highly dependent on conditions: it’s harder to compute a hurricane than a clear blue sky. It would be extraordinarily expensive to build for peak needs in the mainframe paradigm, but this is a use-case well served by cloud computing.</li><li><strong>Low cost sensors. </strong>The dividends of the smartphone wars that have enabled drones and the Amazon Echo will also enable low cost weather sensors, which will create extraordinarily granular weather data. If we had “perfect information” of weather patterns — we should be able to predict it perfectly. We won’t, but we can get much closer.</li><li><strong>Machine learning. </strong>One of the remarkable observations in Nate Silver’s Signal and Noise (2012) is that weather prediction is significantly improved by having expert humans annotate the computer’s prediction. The reason, dated to the book’s publication five years ago, is that computers aren’t very good at reading images. This has has changed. Anything where a human expert is manually applying a set of visual rules on top of a computing prediction is likely highly inefficient.</li></ol><p>But not all weather data is equally valuable. Can you predict air streams to provide better guidance for airline fueling? Can you better target hurricanes evacuations? How much more useful can seasonal weather predictions become for agriculture?</p><p>I suspect the best places to begin are outside of the United States, where you’re not competing against a freely distributed government service, with a budget of billions, and hundreds of years of expertise.</p><p>Now, tell me how I’m wrong.</p><p><strong>If you liked this post, you may like my next one. </strong><a href="https://tinyletter.com/sgerstenzang"><strong>Sign up here to get the next one by email.</strong></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6069c45c9c21" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Questions You Must Not Answer]]></title>
            <link>https://medium.com/@gerstenzang/the-questions-you-must-not-answer-44de4a54f94f?source=rss-996a11626697------2</link>
            <guid isPermaLink="false">https://medium.com/p/44de4a54f94f</guid>
            <category><![CDATA[startup]]></category>
            <dc:creator><![CDATA[Sam Gerstenzang]]></dc:creator>
            <pubDate>Mon, 10 Jul 2017 00:13:09 GMT</pubDate>
            <atom:updated>2017-07-10T00:13:09.011Z</atom:updated>
            <content:encoded><![CDATA[<p>There are some answers that aren’t worth knowing. At least not right now. It might take too much effort to find. The question might be wrong. The world might change. It might be unknowable.</p><p>The traditional analytical approach is to “do the work” to understand what you don’t know, piece by piece. To framework the world and to fill in all the details.</p><p>But I think this approach doesn’t work with big unknowns, or when you’re trying to make big jumps. It doesn’t work in research. It doesn’t work in startups. It doesn’t work in writing.</p><p>Because what you must do is let a new question hang open wide, and use that to go find concrete answers elsewhere. The market is between 1x and 100x its existing size if we reduce the cost by half. We can make this 30x faster but we don’t know who will use it, or what can now be built. Let’s assume we have ten times as much computing power.</p><p>And in exploring, you discover new questions and new answers. Maybe you were wrong but it didn’t matter because the question was the wrong too. Or an adjacent possibility is born, from your own work or from pure luck.</p><p>This suspension of disbelief is critical because our models of the world are very poor. Even if we build all the spreadsheets and all the TensorFlow models, the unmodeled world dominates. We just don’t know.</p><p>The role of the researcher, the essayist, and the businessperson are similar in this way. Success comes from successive suspension of disbelief.</p><p>But to make great things you have to get the ordering right; to hold just enough consistency while letting the pieces wobble themselves out. And of course, not answering some questions will kill you.</p><p>It’s a little bit like living in a house made of jello. You must run from collapsing beam to collapsing beam, massaging each to hope it stays put for a little bit longer. The rain is pouring down through the roof, the neighbor walks by, and you must insist “it’s a house, I swear!” Because if you don’t, it will all come falling down.</p><p><strong>If you liked this post, you may like my next one. </strong><a href="https://tinyletter.com/sgerstenzang"><strong>Sign up here to get the next one by email.</strong></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=44de4a54f94f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Silicon Valley’s Original Sin: The Traitorous Eight]]></title>
            <link>https://medium.com/@gerstenzang/silicon-valleys-original-sin-the-traitorous-eight-b1ba20958cb?source=rss-996a11626697------2</link>
            <guid isPermaLink="false">https://medium.com/p/b1ba20958cb</guid>
            <category><![CDATA[silicon-valley]]></category>
            <category><![CDATA[startup]]></category>
            <dc:creator><![CDATA[Sam Gerstenzang]]></dc:creator>
            <pubDate>Wed, 05 Jul 2017 12:45:29 GMT</pubDate>
            <atom:updated>2017-07-05T12:45:29.824Z</atom:updated>
            <content:encoded><![CDATA[<p>Silicon Valley’s original sin was an act of betrayal. Eight employees left Shockley Semiconductor to found their own company, and William Shockley named them the Traitorous Eight. <br> <br>The newly formed company, Fairchild Semiconductor, was as successful as we can only hope our own ventures are. But it had its own challenges. For one thing, it was funded and owned by corporate East Coast capital. Fairchild Camera and Instrument had provided the initial capital in exchange for an option to outright acquire Fairchild Semiconductor, in what would end up being a short-term bargain for the parent company but severely limited the upside of its subsidiary’s talented team.<br> <br>And so the Traitorous Eight would scatter to form and join their own new ventures, and sow the Valley. Eugene Kleiner founded Kleiner Perkins. Gordon Moore and Robert Noyce founded Intel. Other alum of Fairchild Semiconductor would go on to start their own companies. Jerry Sanders, Edwin Turney, and others founded AMD. Don Valentine founded National Semiconductor and later Sequoia Capital.<br> <br>In the Traitorous Eight, we find the Silicon of Silicon Valley. But we also find the birth of a unique culture, supported by California’s generous employment law, that allowed skills and culture to be spread, absorbed, and reborn, transmitted in the heads of its people.</p><p>In the Valley, skills and techniques are borrowed and cross-pollinated, creating fierce competition for any individual company, but a far richer ecosystem and higher ceilings for the winners. There are three benefits to this cross-pollination.</p><p>The first benefit is that each replanting creates a slightly different context. These re-appropriated cultural contexts allow new types of ideas to flourish, for culture defines what is rewarded and what is possible.</p><p>To take one modern example, the stark differences between Facebook, Google, and Apple’s cultures allow different kinds of products to be built, startups (like Snap) are vehicles for new cultural norms to flourish. At Google, those who ship things that scale are revered. At Facebook, shipping quickly is critical. At Apple, it’s important to take the time to ship the right thing. Doing any of these at another too much and you’ll get fired.</p><p>The second benefit is pure information transfer. It is a legal way to allow technical and process knowledge to seep out, which as the long as the original R&amp;D is rewarded, is a boon to the ecosystem and to its consumers.</p><p>The third benefit is that when high growth companies occur repeatedly, the same kinds of things need to happen over and over again. When the market is less liquid for “inflection point” employees, it is far more costly to find the VP of Engineering who knows how to scale an engineering team from 10 to 50, or the VP of marketing who can stand up the first mass market plan. If the legal and cultural norms reward 10 year tenures, the market is deprived of those career “five to fifty” contributors and executives who make the difference. And of course, there are fewer companies started, period.<br> <br>This talent fluidity is core to ecosystem of the Valley. If the legal or reputation costs of switching firms is high, the ecosystem suffers and the incumbents benefit. Amazingly, the Valley’s Original Sin is still sinful elsewhere in the world.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b1ba20958cb" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[When bandwidth became free, mobile edition]]></title>
            <link>https://medium.com/@gerstenzang/when-bandwidth-became-free-mobile-edition-173149923652?source=rss-996a11626697------2</link>
            <guid isPermaLink="false">https://medium.com/p/173149923652</guid>
            <category><![CDATA[mobile]]></category>
            <category><![CDATA[product-management]]></category>
            <dc:creator><![CDATA[Sam Gerstenzang]]></dc:creator>
            <pubDate>Mon, 29 May 2017 22:00:46 GMT</pubDate>
            <atom:updated>2017-05-29T22:01:03.165Z</atom:updated>
            <content:encoded><![CDATA[<p>One of the most remarkable parts of Snap’s S-1 (<a href="https://medium.com/@gerstenzang/product-lessons-from-the-snap-s-1-dfd3ded1c65c">see my larger analysis here</a>) was the explicit decision to not pursue the ‘lite’ strategy taken by Snap’s social network peers like Facebook and Twitter:</p><blockquote>Our focus on innovative camera experiences means that many of our products are data intensive and work better on high-end mobile devices. This is because camera products involve rich formats like video, which use a lot of cellular bandwidth when used for communication and content consumption.</blockquote><blockquote>Additionally, our products often use technologies that demand a lot of processing power and don’t work as well on lower-end devices, like the technology behind Lenses. This means that unlike many other free mobile applications, the majority of our users tend to be located in markets with high-end mobile devices and high-speed cellular internet.</blockquote><p>In the ‘lite’ strategy, a company takes a product that has worked well in developed markets and adapts it to work in a less developed market with slower phones and more expensive networks: our product doesn’t fit their markets, so we’ll change the product.</p><p>Snap says something different: these markets don’t fit out products so we’ll be less successful there.</p><p>But it was easier for the first generation of mobile apps like Facebook to do this without significant product compromises. The first native Twitter app and the first native Facebook shipped on smartphones approaching a decade old: they were memory constrained, CPU constrained, battery constrained. So going in reverse isn’t too difficult; it’s just optimizing today’s features on yesterday’s platform target.</p><p>The challenge for the next generation of apps (of which Snapchat is an pioneer) is that the newly missing constraint isn’t just the smartphone hardware: it’s the network.</p><p>Snapchat was the one of the first applications to pretend bandwidth was free. And in the last year, we’ve seen more and more apps begin to adopt this assumption. One example: Spotify has begun streaming related songs after a playlist ends. The experience is seamless, but also unthinkable when a megabyte costs you anything you can calculate.</p><p>Another example: my new favorite app <a href="http://gixo.com/">Gixo</a>, which puts you in a live workout class with a real instructor wherever you are. You’d create a very different (and less interesting) product if you had to cache the assets before you went on a run; assuming bandwidth is free enables a new experience.</p><p>The challenge for Snapchat, for Gixo, and for Spotify is that bandwidth isn’t free everywhere yet. And smartphone OEMs move far faster than network infrastructure operators, not to mention you can replace individual smartphone units through consumer demand before replacing the network. The poorest person can afford the richest’s person smartphone hand me down, but not so for their network.</p><p>This means we might see a forking of applications of the next few years. In developed markets, bandwidth is the newly fallen constraint that will enable new kinds of experiences that aren’t possible to translate to developing markets. In developing markets, we’ll see apps that thrive in low bandwidth environments that don’t really make sense when bandwidth is ‘free’.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=173149923652" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>