A common trope in discussions about startups & venture capital is a potential misalignment of incentives between startup team & investors.

Written on March 24, 2015, 10:32 p.m.
  1. A common trope in discussions about startups & venture capital is a potential misalignment of incentives between startup team & investors.

  2. I don't think this perceived misalignment actually exists in most, maybe all cases -- and I want to explain why.

  3. The argument goes, When you take VC, you have to shoot for the moon; smaller outcomes that may be great for the team are precluded.

  4. First, obvious but important: No startup is forced to take venture capital. In fact, vast majority of successful new businesses do not.

  5. The main reason TO TAKE venture capital is in pursuit of a bigger outcome than the startup team believes it could achieve on its own.

  6. Hence, taking venture capital ALIGNS interest in a big outcome between the venture investors and the team.

  7. Most non-VC investors are more risk averse than VCs; a startup shooting for big outcome that raises money from non-VCs ends up MISALIGNED.

  8. The argument cont'd: VCs can tolerate higher risk of failure due to portfolio of bets, whereas founders and employees have only one bet.

  9. In the modern era, that's also untrue. Founders and employees generally make multiple bets as well, in two dimensions:

  10. Founders and employees often have running room to try multiple products within a single startup; hence popularity of the term pivot.

  11. And, founders and employees often start or join multiple startups throughout their multi-decade careers = a personal portfolio of bets.

  12. In fact, some of today's most successful startups are founded by entrepreneurs whose previous ventures didn't work nearly as well.

  13. In short: VC is very much not for every company. But for companies that want to do something big, VC = the most aligned capital there is.

  14. 3 of people in the Valley who have completed their college education are foreign born. https://t.co/Ed5L73laSi

  15. California's datacenters consume 158,000 Olympic sized swimming pools of water per year, or 104,280,000,000 gallons of water per yaer.

  16. That sounds like a lot. Enough, in fact, that one might wonder what else might consume that much water every year.

  17. It turns out that a single day's supply of newspaper newsprint requires about 300,000,000 gallons of water. https://t.co/IvvjEKuZbD

  18. Multiply by 365 days per year, producing the US's annual consumption of newsprint consumes 109,500,000,000 gallons of water per year.

  19. 109,500,000,000 gallons per year for US newsprint > 104,280,000,000 gallons for California datacenters. Interesting!

  20. Of course newsprint consumption is falling due to substitution by online media! And we know online news is a very small % of online use.

  21. So, for a small fraction of the water use by datacenters, over time we can likely eliminate nearly all of the water use for newsprint.

  22. Not to mention reducing the deforestation caused by newsprint production. Less tree use, less water use = environmental win/win!

  23. It's interesting to consider what other current environmental resource uses will be reduced or eliminated by datacenters & online media!

Sharing economy services like AirBnB, Lyft, Uber, et al reduce income inequality, as follows:

Written on March 19, 2015, 1:02 a.m.
  1. Sharing economy services like AirBnB, Lyft, Uber, et al reduce income inequality, as follows:

  2. Once upon a time, only rich people who could afford to build hotels could offer rooms for guests to rent.

  3. Thanks to AirBnB, now anyone with a home or apartment can offer a room for rent. Hence, income inequality reduced.

  4. Once upon a time, only rich people who could afford to buy a taxi fleet or medallion could offer car rides to other passengers.

  5. Thanks to Lyft and Uber, now anyone with a car can offer rides to other passengers. Hence, income inequality reduced.

  6. The same dynamic of reduced income inequality will play out in each new sharing category. Delivery, field service, personal care, etc.

  7. This expansion of economic oppy and reduction of inequality is direct result of widespread deployment of disruptive tech: the smartphone.

  8. As history has repeatedly shown, putting means of production - technology - in the hands of the masses increases their oppy and income.

  9. And since we are at the very beginning of understanding the full range of real world applications of the new wave of technology...

  10. We are also at the very start of the creation of thousands of new opportunities for economic growth and progress for huge #s of people.

In financial markets, one observes two different types of relationship between prices and information.

Written on March 4, 2015, 1:36 a.m.
  1. In financial markets, one observes two different types of relationship between prices and information.

  2. Type D for Deductive: Consider the available information and then calculate a price. The classical model of how to value things.

  3. Type I for Inductive: Consider the price, assume it contains information content, and derive the information from the price.

  4. In theory financial markets operate mainly with Type D, but I think in practice markets operate mainly with Type I.

  5. In the real world, one observes investors and analysts assiduously building models to explain and thereby justify prevailing prices.

  6. Paradox? The more one believes the market is Type D, aka the EMH, the more the market actually behaves as Type I.

  7. The more you believe the market is efficient, the more information you assume is embedded in market prices.

  8. Hence, the more information you assume is embedded in market prices, the more you operate as Type I, inductively reasoning from prices.

  9. Therefore, the more one believes the rest of the market is Type D, the more likely oneself is Type I.

  10. Hence widespread belief in market efficiency leads to inefficiency, as investors reason from prices vs from a priori information?

  11. Is this a robust explanation of boom/bust cycles within a market in which most investors are trying to be rigorously logical?

Idle Friday musing...unlike many of my contemporaries, although I enjoyed space novels & movies as a kid, it didn't take the same with me.

Written on Feb. 27, 2015, 10:44 p.m.
  1. Idle Friday musing...unlike many of my contemporaries, although I enjoyed space novels & movies as a kid, it didn't take the same with me.

  2. The common science fiction assumption that people would behave differently in space than they do on Earth always struck me as wrong.

  3. Which is why I'm delighted but not enamored with space exploration today. There is so much still yet to do here on Earth.

  4. I think we will get a lot more out of making life on Earth better for the 7+ billion people already here than escaping to other planets.

  5. Or, more pointedly, it would be shame if we just repeat the dysfunctions of human existence on Earth, in space.

  6. Which is not to imply that I am not thrilled by what explorers like Elon Musk are doing. I would love to visit Mars when I am 80 :-).

  7. Confirming my nerd status, I have happily watched every hour of Star Trek ever made, and so: Live long and prosper, Leonard Nimoy.

  8. 07: We Americans: noses pressed against shop window of the future, admiring technology we only dream of owning. http://t.co/JzJrIRK3ZY

More Montessori & Montessori-style, free-form, and/or project-based K-8 public & private schools.

Written on Feb. 16, 2015, 5:31 a.m.
  1. More Montessori & Montessori-style, free-form, and/or project-based K-8 public & private schools.

  2. Entrepreneurship magnet/charter schools -- specifically designed to produce enterpreneurs, vs cogs in the industrial machine.

  3. Significantly expanded summer tech, science, math, entrepreneurship programs/camps for grades 5-12.

  4. Significantly expanded internship programs at tech companies of all sizes for both high school and college students.

  5. More interdisciplinary college programs -- particularly engineering + business, and liberal arts + engineering.

  6. Comprehensive inclusion programs for underrepresented groups for each of the preceding five ideas.

  7. More public research universities should pursue the Stanford/Berkeley mentality/model; also, repeal Bayh-Dole.

  8. Comprehensive legal and regulatory reform to open access to federally-funded research; also, pass Aaron's Law.

  9. Reform, or better yet eliminate, software and business method patents. Redefine patent trolling as a form of felony extortion.

  10. Fully portable economy-wide benefits, including health care, retirement savings, and immigration status.

  11. Eliminate tax credits for home ownership, and implement tax credits for renters.

  12. Implement tax credits for child care services for working parents.

  13. Opt in innovation zones with regulatory relief for various categories of new technology.

  14. More long-lockup capital at all levels of corporate capital structure.

  15. Eliminate tax credits for corporate debt, and implement tax credits for corporate equity.

  16. Zero capital gains tax for equity held for 5+ years, paid for by higher capital gains tax for equity held for <2 years.

Some thoughts on the state of Bitcoin, cryptocurrencies, and distributed transaction and trust networks at start of 2015!

Written on Jan. 5, 2015, 7:33 p.m.
  1. Some thoughts on the state of Bitcoin, cryptocurrencies, and distributed transaction and trust networks at start of 2015!

  2. A year ago I articulated my views on Bitcoin in the New York Times, and I wouldn't change a word today: http://t.co/2yuRx724ZJ

  3. Over the last several months of 2014, there were three Bitcoin counter-narratives that I will describe and analyze:

  4. First, what I call the dumb critique: Bitcoin BTC currency price dropped to below $300, proving Bitcoin is a stupid idea.

  5. Same class of critics slamming BTC for price falling in 2014 were slamming BTC for price rising in 2013. Only consistency is the slamming.

  6. Further, the critique that BTC is bad because it was down in 2014 changes completely if one uses a 2-year window instead of 1-year window.

  7. BTC was below $14 in Jan 2013. If 1-year performance has been disappointing, 2-year performance has been spectacular.

  8. As a general rule, arguments that rely on cherry-picking specific date windows are not very good arguments.

  9. The second critique I call smarter: BTC is too volatile -- it goes up and down too much and so cannot be used as a store of value.

  10. This is largely correct at the moment, and yet misses most of the point of Bitcoin as a distributed transaction and trust network.

  11. Bitcoin was specifically designed to use speculation early on to overcome the normal chicken/egg boostrapping problem for new networks.

  12. Attacking Bitcoin for having speculative levels of volatility is missing the point of how the system was designed for this point in time.

  13. Now, yes, in the long run, BTC does need to stabilize -- which I think will happen with a combo of scale + use of derivatives (hedging).

  14. In the short run, Bitcoin is still highly useful as a transaction and trust network in many uses cases even with high volatility.

  15. For example, payment applications of BItcoin don't require users/merchants to hold BTC for any period of time. All benefits still gained.

  16. Further, all other uses cases of Bitcoin and the blockchain are unhampered by volatility of BTC. The system continues working just fine.

  17. Net net, the network boostrapping process is happening pretty much exactly as Satoshi designed and anticipated. It's a thing of beauty.

  18. The third critique I call the innocent one -- Are there enough sufficiently compelling uses cases for Bitcoin to succeed at scale?

  19. I previously identified many uses cases here http://t.co/UHQL5HQQ2O ranging from ecommerce to remittance to micropayments to anti-spam.

  20. Two particular areas of focus today are A use outside the US where currencies + banks are often awful, and B machine-to-machine payments.

  21. At our venture firm, we continue to see an escalating stream of fascinating new Bitcoin uses cases and applications from entrepeneurs.

  22. In addition, there are entirely new vistas of technological creativity opening up, such as sidechains http://t.co/KLLKVtr6dE.

  23. The price of BTC has very little to do with the level of creativity of thinking that's going into new Bitcoin apps, or their usefulness.

  24. By loose analogy, the price of domain names didn't determine the usefulness of the Internet. This is a broad-based technology phenomenon.

  25. What to watch in 2015: New apps, new use cases, international adoption, consumer education, technological innovation & spinoff ideas!

  26. Final thought: The entire Bitcoin system is 6 years old. TCP/IP was 6 years old in 1981. Big things take time. Onward!

Wrapping up my series on the secular stagnation theory with some more things I learned along the way. Again, filter on #stag as desired.

Written on Dec. 30, 2014, 9:44 a.m.
  1. Wrapping up my series on the secular stagnation theory with some more things I learned along the way. Again, filter on #stag as desired.

  2. At some point Europe is going to either need a lot more immigrants or a lot more robots: http://t.co/WBN0JhdkBw

  3. Edward Glaeser: Who would really be indifferent between earning $23K in 1984 vs $50K in 2014? #stag http://t.co/JuTNptO2BH

  4. Glaeser: Before modern industrial capitalism, innovation and trade mainly benefitted the elite. #stag http://t.co/gqiQNtB1K6

  5. Glaeser: Startling rise in reported disabillity rates among working-age Americans over last 50 years. #stag http://t.co/lpHZifVm2H

  6. Glaeser: 72.7% of US college grads ages 25+ are employed, vs 39.4% of high school dropouts in same cohort. #stag

  7. I'd like to quote Mokyr's entire essay in the Vox ebook but that would be rude, so read the whole thing here: http://t.co/KUAhYLuUxM #stag

  8. Nicholas Crafts: Boy, is Europe in trouble. #stag http://t.co/7ggNJjKCCx

  9. Nicholas Crafts: Europe badly needs to step up technology adoption and productivity growth. #stag http://t.co/KEejDuML1D

  10. Olivier Blanchard: Post-crisis reforms require financial institutions to hold another $3 trillion in safe assets. (!) #stat

  11. Blanchard: China savings pattern is opposite of normal assumption, due to poor social insurance programs. #stag http://t.co/Fv5o4ZHun4

  12. Koo: Those who prevent crises never become heroes, so at least give the impending threat a great name! #stag http://t.co/dBwZifjZX0

A few tentative conclusions on secular stagnation & our economy, with vigorous disclaimer that I am far from a macroeconomist... #stag

Written on Dec. 29, 2014, 11:15 a.m.
  1. A few tentative conclusions on secular stagnation & our economy, with vigorous disclaimer that I am far from a macroeconomist... #stag

  2. It seems a core dynamic of our times is too much capital relative to the number of productive investible economic opportunities. #stag

  3. Coupled with a massive global capital flight to quality since 2008, hard to see interest rates rising dramatically anytime soon. #stag

  4. While I am a bull on technological progress, it also seems that much of that progress is price deflationary in nature...

  5. ...so even extremely rapid tech progress may not show up in GDP or productivity stats, even as it = higher real standards of living. #stag

  6. I think economists, particularly on the center-left and left, are really underestimating 2 factors that are inhibiting investment. #stag

  7. In developed world, sheer level of regulatory burden on business formation and growth. Per George McGovern: http://t.co/jqEQdPiWgp #stag

  8. In the developing world, often brutally high levels of corruption and expropriation, making new investment extremely risky. #stag

  9. It seems straightforward to identify ways to increase rate of investment, and also hard to see how any of that politically happens. #stag

  10. For these and other reasons, we may be living with an oversupply of capital relative to opportunity set for a long time. #stag

  11. But this is not necessarily a terrible world to live in. In fact, it might be a wonderful world to live in, for these reasons: #stag

  12. Oversupply of capital means that any investable project can get funded. We see that today in tech, and it may broaden from here. #stag

  13. We may experience a massive global demographic tailwind, as huge # of young people worldwide are fully connected to modern economy. #stag

  14. Virtuous cycle of science & tech advances, with fast-growing # of scientists & technologists globally, may overwhelm expectations. #stag

  15. In this world, we can have massive advances in real standards of living even w/formally low investment, GDP, & productivity growth. #stag

  16. Beyond that, a world where 7 billion people decide they really do want and deserve an upper-middle-class Amercan-equiv lifestyle... #stag

  17. ...may make all of these current stagnation theories look as silly as Alvin Secular Stagnation Hansen now looks 76 years later. #stag

Thinking about Larry Summers' secular stagnation thesis -- if the topic bores you, filter out hashtag #stag for a bit :-). #stag

Written on Dec. 29, 2014, 4:43 a.m.
  1. Thinking about Larry Summers' secular stagnation thesis -- if the topic bores you, filter out hashtag #stag for a bit :-). #stag

  2. Contrary to a lot of public discussion, Larry's thesis is about interest rates and supply/demand of capital, not technology change. #stag

  3. Core to secular stagnation is substantial decline in natural real rate of interest. #stag http://t.co/kW9AIMrxZC http://t.co/uKkoRb7zU1

  4. Factor 1: Reduced demand for debt-financed investment, in part due to changing nature of economic activity. #stag http://t.co/E3gbtkqQaJ

  5. Factor 2: Declining rate of population growth, especially if one corrects for education levels. #stag http://t.co/VEZGM8RrOu

  6. Factor 3: Increased societal propensity to save + Increased corporate retained earnings. #stag http://t.co/RXrCFD2kck

  7. Factor 4: Decline in relative prices of business & consumer capital goods (tech-driven price deflation?). #stag http://t.co/AY2PKVkoFD

  8. Factor 5: Arithmetic relationship between lower pre-tax and after-tax real interest rates. #stag http://t.co/oqh5vMuwb2

  9. Factor 6: Growing central bank reserves of safe assets, particularly US Treasuries. #stag http://t.co/ATM7tGCjz7

  10. Result: Substantial and continuing decilne in real interest rates over 50 years. #stag http://t.co/8o30RkZkg9

  11. To quote Notorious BIG, mo money mo problems -- in this case, mo money, not enough productive places & projects to put it. #stag

  12. As economists do, Larry then proposes a series of reforms to address the dynamic, few of which seem politically likely. #stag

  13. References: http://t.co/kW9AIMrxZC http://t.co/KUAhYLuUxM http://t.co/wma6y5P1TQ #stag

  14. Lingering question from Larry Summers' secular stagnation thesis: #stag http://t.co/8p66PvNWfk

  15. Doesn't the same question apply for the past 100 years? Has the American economy ever met that criteria? #stag http://t.co/HtHXpB9FPS

  16. Can the goal be to get to normal if there is no normal? Or should the goal be something else? #stag

  17. First point: Prior risk premium in bonds warped prices+yields, made credit artifically expensive in the past. #stag http://t.co/EbJwpDYydq

  18. Risk prem boosted by high inflation, so lower inflation reduced rate TWO ways, lower inflation & lower risk. #stag http://t.co/Mn20krtbDS

  19. Decomposing interest rate from underlying inflation expectation + risk premium; both came down since 80s. #stag http://t.co/9ynJZvr3Zd

  20. With low risk of future extreme inflation, real interest rate bounces along around 2%, as it should. #stag http://t.co/5w3yXj5AN6

  21. Result: Secular stagnation vanishes, along with misbehaving interest rates. #stag http://t.co/qTtvXkpnn8

  22. Further, productivity growth may not be as bad as we think, in era of software/digital/dematerialization. #stag http://t.co/ZmssWFVf0z

  23. Finally, the millennials are coming--both in the US and around the world--tilting demographics in our favor. #stat http://t.co/9m3WadSsTG

  24. In this interpretation, there is no secular stagnation, there was just a crisis and then a slump, which is already ending. #stag

  25. References: http://t.co/moDmYn4sJs http://t.co/hBslYBBIMC http://t.co/ITGbwZbQ7z http://t.co/rh3ModRiki http://t.co/fr6ImW3UI1 #stag

  26. Hansen's 1938 thesis will sound familiar to post-2008 ears: #stag http://t.co/I2FctorPLw

  27. We know now Hansen in 1938 was thoroughly wrong. For example, he totally whiffed on population projections. #stag http://t.co/MklTZhxve1

  28. Hansen was also completely wrong on the rate of technological progress at that time. #stag http://t.co/vVxV7RsyHx

  29. However, he did issue a ringing call for new technologies and new industries that applies equally to today. #stag http://t.co/I7bLm5bO5G

Considering the Bernanke global savings glut hypothesis from 2005 -- what holds up, what doesn't: http://t.co/ZHXTSIdvkj

Written on Dec. 26, 2014, 8:34 a.m.
  1. Considering the Bernanke global savings glut hypothesis from 2005 -- what holds up, what doesn't: http://t.co/ZHXTSIdvkj

  2. Ah, the good old days of 2005, when the major US macroeconomic concern was the trade deficit: http://t.co/A1lsv38viF

  3. The global savings glut thesis: http://t.co/T8OiLxYjWc

  4. Global savings glut in rich countries with aging populations, which run trade surpluses and invest abroad: http://t.co/Xy8UtRLKNd

  5. The role of financial crises in causing developing world countries to invest more abroad: http://t.co/iwzIexiF1J

  6. The global savings glut after the 2000 equity market crash, and the low interest rates that followed: http://t.co/eYsV2AanGG

  7. The oddness of developing countries investing in developed countries rather than the other way around: http://t.co/DqNsxgNtLI

  8. Getting to the heart of what I suspect is the actual underlying problem: http://t.co/IfxPLsseKG

  9. There it is: http://t.co/h7ZYPrNeI4

  10. On global current account deficits and surpluses after Bernanke 2005 and the 2008 financial crisis: http://t.co/1KcZGCVSPS

  11. See page 199 for updated current account balances by country and region through 2013: http://t.co/Q795Sq6cZ0

  12. Even after developed world financial crisis, developing world still invests much more in developed countries than vice versa.

  13. Like Bernanke, I believe in Wriston's Law: Capital will go where it is wanted and stay where it is well treated. http://t.co/sW40jeY5Li

  14. How would global capital flows look if all countries treated capital as well as the US does? (Not that we lack issues, to be sure.)

Over the last 2 days I've been tweetshotting excerpts from David Wells, Recent Economic Changes, 1890: https://t.co/814vpfqcBH

Written on Dec. 23, 2014, 5:44 a.m.
  1. Over the last 2 days I've been tweetshotting excerpts from David Wells, Recent Economic Changes, 1890: https://t.co/814vpfqcBH

  2. For background reading on the economic period of ~1870-1890: http://t.co/EPbpyeRK8H http://t.co/zGLTrtpCBA

  3. This period is not a direct historical analog to ours, but it's probably as close an analog as there is, along with the 1920s-1930s.

  4. Each of the three periods struggled with serious macroeconomic crises, albeit substantially different natures. Hard to compare those.

  5. People of that time did wonder the same questions that are being so frequently asked today:

  6. What is the nature of technology-driven economic change and creative disruption?

  7. What is the future of income and wealth distribution and returns from progress -- for labor and for capital?

  8. How do present changes compare to those in the past, and what can be forecast about the changes yet to come?

  9. What kind of world are we building, and what kind of world will we leave for our children and grandchildren?

  10. Wells, writing in 1890, would have been wholly gobsmacked at the widely distributed gains of the next 100 years! http://t.co/5mE6cUfYlF

  11. It may not surprise you that I think we are going to repeat what Brad describes in http://t.co/5mE6cUfYlF in this century again.

  12. But, we actually have to prove it, and do it.

Fascinating overview of the secular stagnation hypothesis from the Bank of Italy (in English): https://t.co/rfnHkupv2o

Written on Dec. 19, 2014, 11:50 a.m.
  1. Fascinating overview of the secular stagnation hypothesis from the Bank of Italy (in English): https://t.co/rfnHkupv2o

  2. After reviewing recent long-run projections, we argue that similar warnings were issued in the past after all deep recessions.

  3. Interestingly, pessimistic predictions turned out to be wrong neither because they were built on erroneous theories or data...

  4. ...nor because they failed to predict new tech, but because they underestimated the potential of the technologies that already existed.

  5. These findings suggest that today we should not make the same mistake and undervalue the effects of information technology.

  6. This matches my personal belief: Much current economic commentary is result of living through a 15 year down cycle, which will change.

  7. Interestingly, our friend Larry Summers on CNBC today conceded secular stagnation may be more a Europe/Japan issue than a US issue.

  8. If the US economy is indeed at the front of a broad-based recovery, as it appears, it will be interesting to see where this issue lands.

Utah attempting to ban Zenefits = classic example of regulatory capture penalizing consumers to benefit incumbents. http://t.co/fcd52e53VR

Written on Dec. 3, 2014, 12:52 a.m.
  1. Utah attempting to ban Zenefits = classic example of regulatory capture penalizing consumers to benefit incumbents. http://t.co/fcd52e53VR

  2. This action by Utah is a direct friendly-fire strike on Utah's small businesses, and Utah's own ambitions as a fertile high-tech hub.

  3. In fact, Mixpanel is unusual in that the new investment round will not go to accelerating growth, opening int'l offices, or the like.

  4. Mixpanel generates cash internally to fully fund continued expansion of its current business. Rather, Suhail has more ambitious goals.

  5. We originally invested in Mixpanel because it makes state-of-the-art mobile/web analytics easy for every company in the world.

  6. Mixpanel lets you track actions not just pageviews (The smartest entrepreneurs pitching us were showing us their data thru a MP dashboard)

  7. The broader goal: Help the world learn from its data -- to bring data science to every domain, to fundamentally improve how things work.

  8. Mixpanel will use the new financing to build new products, acquire, break into new markets -- and take crazy risks that may work or fail.

  9. Mixpanel will go straight after the goal of predicting the future with data -- what we think is the next phase of analytics.

  10. And always, Mixpanel will be about merging the best of what machines can do with the best of what people can do: processing + judgment.

Few intellectual concepts in our time have been mangled by observers more than Clay Christensen's disruption idea. Some thoughts:

Written on Dec. 2, 2014, 4:56 p.m.
  1. Few intellectual concepts in our time have been mangled by observers more than Clay Christensen's disruption idea. Some thoughts:

  2. CC: A disruptive innovation gives new consumers access to product historically only available to consumers with a lot of money or skill.

  3. CC: Disruptors offer a different set of product attributes valued only in new markets remote from, and unimportant to, the mainstream.

  4. The key attribute of disruptive innovation is a new product for a previously underserved market--typically cheaper than existing product.

  5. This is inherently pro-consumer: Disruptive innovation only works if customers buy it--and if they do, lives improved vs prior status quo.

  6. Similar, disruptive innovation is only funded by investors who believe underserved market exists, customers will buy it, lives improved.

  7. It's a fabricated myth that disruptive innovation is about destruction: It's about creation--new products, new choices, for more people.

  8. Later, of course, new product often evolves to squarely take on incumbents serving established customers--cheaper & better for them too!

  9. Disruptive innovation shrinks inequality, by bringing to lower-income consumers things that only richer consumers had access to before.

  10. If you are reading this, many of the things you own that make your life better are the result of prior disruptive innovation.

  11. Printing press disrupted books from scribes; recorded music disrupted live concerts in homes, washing machines disrupted live-in maids.

  12. Rich people always had books, music, clean clothes, etc.; disruptive innovation made these things available to many more people.

  13. In exact same way, sub-$50 smartphones as disruptive innovation to PCs bringing computing & Internet to far more people than status quo.

  14. To be FOR disruption is to be FOR consumer choice, FOR more people bring served, and FOR shrinking inequality.

  15. To be AGAINST disruption is to be AGAINST consumer choice, AGAINST more people bring served, and AGAINST shrinking inequality.

  16. If we want to make the world a better and more equal place--the more Christensen-style disruption, and the faster, the better!

  17. References: http://t.co/9dbASWMYgt, http://t.co/IZnvsmiQ71, http://t.co/9L6ZKjiAgS

Hedge fund baron Paul Singer: London, Manhattan, Aspen, & East Hampton real estate, & art, prices [show] leading edge of hyperinflation.

Written on Nov. 14, 2014, 5:57 a.m.
  1. Hedge fund baron Paul Singer: London, Manhattan, Aspen, & East Hampton real estate, & art, prices [show] leading edge of hyperinflation.

  2. There are a bunch of reasons to believe that his theory is wrong, but one that I think is under-appreciated is this:

  3. Those are specific markets seeing dramatic influx of ultra-high-net-worth buyers from overseas--China, Russia, and certain other nations.

  4. You can see this vividly in this week's art auctions in New York, and it's been obvious in London real estate for some time.

  5. Another: 30% of all apartments 49th-70th Streets between Fifth&Park are vacant at least 10 months a year. http://t.co/WkMCc0fIG3

  6. So price rises in these specific assets can likely be explained by simple supply & demand without requiring inflation, hyper or otherwise.

  7. And therefore price rises in these assets do not necessarily indicate much of anything about the domestic macroeconomic situation.

  8. In fact, obvious bullish on America argument here: Capital fleeing other countries and landing specifically in US (& UK!) real estate.

  9. Further, these asset prices may be explainable quite independently of the various QE & inequality debates happening within the US.

  10. On a global economic scale, the total value of this specific real estate + art is not that large. Prices easily altered by capital flows.

  11. Over the past two decades, what have the US trends been for the following important measures of social health:

  12. High school dropout rates, college enrollment, juvenile crime, drunken driving, traffic deaths, infant mortality, life expectancy...

  13. Per capita gasoline consumption, workplace injuries, air pollution, divorce, male-female wage equality, charitable giving...

  14. Voter turnout, per capita GDP and teen pregnancy?

  15. The answer for all of them is the same: The trend is positive. Almost all have improved by more than 20% over the past two decades.

  16. Many Americans are convinced 'half marriages end in divorce': not the case since early 80s--since, they have declined by almost a third.

  17. Even though the world's population has doubled over past 50 years, percentage living in poverty has declined by 50% over that period.

  18. Positive trends in our social health are coming from a complex network of forces [vs big-bang tech breakthroughs].

  19. No one takes out a prime-time ad campaign to tout the remarkable decrease in air pollution that we have seen over the past few decades.

  20. Curmudgeons, doomsayers, utopians & declinists all have easier time getting PR than those who celebrate slow & steady improvement.

  21. In long run, media bias against incremental progress may be more damaging than any bias the media display toward left or right.

  22. The media are heavily biased toward extreme events, and they are slightly biased toward negative events...

  23. ...though in their defense, that bias may be reflection of human brain's documented propensity to focus more on negative information.

  24. We underestimate the amount of steady progress that continues around us, and we misunderstand where that progress comes from.

  25. We should celebrate stories of progress, not to rest on our laurels but so we can inspire the next generation to build on that success.

Enough people asking what I think of net neutrality so I will attempt to answer, but warning, I do not have a clean and simple answer!

Written on Nov. 12, 2014, 6:26 a.m.
  1. Enough people asking what I think of net neutrality so I will attempt to answer, but warning, I do not have a clean and simple answer!

  2. I think permissionless innovation, nondiscriminatory nature of Internet is of critical importance and must be maintained or strengthened.

  3. I also think telco/cable companies need incentives to build far more/better net infrastructure than we have now & be able to make $ on it.

  4. Due to the economics of network businesses, I think these are extremely difficult principles to reconcile & I don't envy the regulators.

  5. I further worry about simplistic and politicized nature of much of the debate, which I think is not conducive to navigating complexity.

  6. And I further still worry about the sausage-making of any regulatory process & the likelihood of unanticipated & undesirable outcomes.

  7. And generally, I try to spend my time trying to figure out how to bring more/better/faster Internet to more people in new/different ways.

Title II FCC regulations were created in 1934, based directly on railroad regulations from 1887, 125+ years ago. http://t.co/auLvBeIY7a

Written on Nov. 12, 2014, 5:58 a.m.
  1. Title II FCC regulations were created in 1934, based directly on railroad regulations from 1887, 125+ years ago. http://t.co/auLvBeIY7a

  2. Here are some other things that happened in 1887, same year railroad regulations that then became telecom regulations were designed:

  3. In 1887, Arthur Conan Doyle's detective character Sherlock Holmes makes his first appearance, in the novel A Study in Scarlet.

  4. In 1887, the earliest constituent of the U.S. National Institutes of Health is established in Staten Island as the Laboratory of Hygiene.

  5. In 1887, construction of the iron structure of the Eiffel Tower starts in Paris, France.

  6. In 1887, Gottlieb Daimler, the cofounder of the company that now makes Mercedes cars, unveils his first automobile.

  7. In 1887, the cornerstone of the new Stanford University, in northern California, is laid (the college opens in 1891).

  8. In 1887, the Giuseppe Verdi opera Otello premieres at La Scala.

  9. In 1887, Emile Berliner is granted a patent for his Gramophone, the flat-disc phonograph.

  10. Finally, in 1887, Chester Greenwood patents earmuffs.

  11. These and many more described here: http://t.co/GZcTOQH5FO

People and studies I referenced onstage at #WSJDLive today:

Written on Oct. 28, 2014, 4:47 p.m.
  1. People and studies I referenced onstage at #WSJDLive today:

  2. Economist Carlota Perez & her book Technological Revolutions and Financial Capital -- http://t.co/OeGrhmo6ez

  3. Economist Claudia Golden & her paper A Grand Gender Convergence: Its Last Chapter -- http://t.co/r1VczsPzHO

  4. Other people and studies I would have mentioned if I had more time:

  5. OECD inspired by economist Angus Maddison: Global Well-Being Since 1820 -- http://t.co/DnZoCEogSn

  6. Economist Julian Simon and his book Ultimate Resource II -- http://t.co/2p9lV9iw6T

  7. That last is underappreciated when considering the long-run growth consequences of wiring everyone on the planet into a single network.

  8. Dr Bucks is in quarantine in his home in Redwood City and is Ebola symptom free: http://t.co/OLtD9ecWBF

  9. Between 26 and 32 patients died during his stint there, yet nearly as many survived because of the care provided at the 52-bed clinic.

  10. Dr Bucks called the Liberian workers at the clinic 'my heroes' because of their determination in the face of daily threat of disease.

After 6 years of service + completed decision to split, I've decided now is a good time to step off the eBay Board of Directors.

Written on Oct. 20, 2014, 9:05 a.m.
  1. After 6 years of service + completed decision to split, I've decided now is a good time to step off the eBay Board of Directors.

  2. It's been an absolute privilege to serve with John, Pierre, and team, and I could not be more proud of what we've accomplished.

  3. I wish eBay, and both of its successor companies, all the best in the years to come & will seek to continue to help as much as I can.

  4. Official announcement: http://t.co/R4wZJwazsu

Given extremely low price/earnings ratios for big public tech companies including GOOG and AAPL, only two outcomes are possible:

Written on Oct. 8, 2014, 6:55 p.m.
  1. Given extremely low price/earnings ratios for big public tech companies including GOOG and AAPL, only two outcomes are possible:

  2. One: Big tech is undervalued, in which case there's no tech bubble. You don't see multiples this low in any bubble.

  3. Other: Big tech is fairly valued, meaning all doomed within ~10 years, therefore their startup disruptors are undervalued, so no bubble.

  4. Our donation was for both the current & a future new Stanford Emergency Department. Today I want to provide an update on the new ED!

  5. Stanford is building an entirely new hospital in the heart of Silicon Valley, which will start operation in 2018: http://t.co/3Q5fmMkbjf

  6. The overall project is 824,000 square feet, and will include our new ED, which will be twice the size of today's and fully cutting-edge.

  7. The hospital will cost $1.8 billion in total. Of that, $600 million in philanthropic donations has already been raised, of a $700M target.

  8. Philanthropic donors include both individuals like us + local companies: Adobe, Apple, Cisco, eBay, HP, Intel, Intuit, Nvidia, and Oracle.

  9. The new Stanford Hospital will be profoundly transformative for the Silicon Valley community, from complex cancer care to 3AM emergencies.

  10. And the new Stanford ED -> Even more powerful healthcare safety net for our lowest-income, homeless, and undocumented immigrant neighbors.

  11. Stanford Hospital CEO Amir Dan Rubin & colleagues are building future of healthcare in Silicon Valley & Laura + I are thrilled to support.

  12. For more information: http://t.co/UMAkk9wZC7 http://t.co/jUqiBe4ROS http://t.co/jf8Ez3k3rk

  13. If you'd like to join us in supporting the new Stanford Hospital, donations can be made online here :-): https://t.co/b4quZez9x9

  14. She says Silicon Valley has the microphone. More people should step up and seize it. -- I wholeheartedly agree!

  15. In that spirit: Here are a bunch of tech people who aren't widely famous (yet) but who routinely say interesting and provocative things!

  16. Most of these folks are not affiliated with me/A16Z -- what they have in common is, when they say things, I learn things!

  17. Of course this is only a highly abridged selection; for several hundred more, see my Industry list: https://t.co/nsEwqOFDlJ

Bloomberg Markets was nice enough to publish an interview with me, mostly on finance and banking topics, this week: http://t.co/ZaQPIhqehY

Written on Oct. 7, 2014, 8:41 p.m.
  1. Bloomberg Markets was nice enough to publish an interview with me, mostly on finance and banking topics, this week: http://t.co/ZaQPIhqehY

  2. Since the interview is correct but abridged, and took place in early April and is just coming out now, I offer a few additional thoughts:

  3. Unbundling banks: I think banks are getting unbundled with or without Silicon Valley or Bitcoin--due to market changes + regulation.

  4. Any big bank executive will tell you: Over time, ratio between what non-banks can offer vs what banks can offer is steadily increasing.

  5. Post-crisis reforms like Dodd Frank are accelerating the unbundling, whether that's what regulators intended or not.

  6. But Dodd Frank is double-edged sword: Also makes it harder for new entrants to the core banking system. Banks both protected & restricted.

  7. Bitcoin and regulation: Since interview in early April to now, there's been substantial movement by many regulators on Bitcoin.

  8. Plenty of current discussions and disputes between market participants and regulators, but overall I think a lot of progress is happening.

  9. Finally, much like the Internet 20 years ago, Bitcoin as a technology can and will be adopted by both incumbents & new market entrants.

  10. We are seeing a rapidly escalating level of engagement and interest in Bitcoin and cryptocurrency by large financial services companies.

  11. The opportunity is clear and present for both big companies and startups to use new technology to improve financial services broadly.

  12. Finally (again :-): Since the interview, the other huge earthquake to hit financial services industry is the launch of Apple Pay.

  13. Between Apple Pay & Bitcoin, I predict more changes coming in financial services & banking in the next 3 years than in the last 20 years.

  14. These three Inclusion Grants total $500,000, and go to three stellar young nonprofit organizations working to grow inclusion in tech:

  15. Laura and I couldn't be more excited to work with all three amazing groups to help them scale their programs over the next several years.

  16. We think there's huge opportunity to include more people, including/especially underrepresented groups, into tech field and tech industry.

  17. Finally, big thank you to Queen Elizabeth Prize for Engineering (http://t.co/5hFjpVH2Bd)--my 2013 prize money helped fund these grants.

One response to https://t.co/MIUJtDFAVz: Why isn't this just hypocritical VCs overfunding reckless founders of out-of-control startups?

Written on Sept. 26, 2014, 4:08 p.m.
  1. One response to https://t.co/MIUJtDFAVz: Why isn't this just hypocritical VCs overfunding reckless founders of out-of-control startups?

  2. In fairness, there is probably some of that, though we & the investors we respect try hard not to indulge recklessness & irresponsibility.

  3. But while it's irresponsible to vaporize cash & your company, it can also be irresponsible to NOT invest to become #1 in a big new market.

  4. Particularly now, since there are SO many more people on the Internet & SO many more businesses that can consume cloud/SAAS vs 15 yrs ago.

  5. Tension: Overinvest, escalate burn, risk down round, vaporize when market turns; OR Underinvest, starve growth, don't win market, implode.

  6. Why is this so important? In tech-driven markets, overwhelming economic returns tend to go to the company with the highest market share.

  7. And, the winning company with the highest market share can invest the most in R&D, build the best and most advanced products. The prize.

  8. Via Glengarry Glen Ross: Reward for market position #1 is 90% of the economic value. #2, a set of steak knives. #3, you're fired.

  9. The challenge for CEOs and boards of tech startups is to thread the needle, just enough investment to take the #1 position, but not more.

  10. Meeting this challenge has resulted in thousands of venture-capital-backed companies creating millions of jobs over last 50 years.

  11. This challenge becomes far harder when money is flowing freely, since more competitors get funded. Very tricky. Requires deep judgment.

  12. BUT opting out of the race generally guarantees you won't be #1 or even #2. Not a good idea either. Just as serious a risk as blowing up.

  13. No single answer. Up to VCs, CEOs, boards, and later-round investors to think very carefully about this for each specific circumstance.

Next generation movie theaters could be so much better, charge a premium & dominate financially (like http://t.co/SnJsIk8528)...

Written on Aug. 31, 2014, 10:35 p.m.
  1. Next generation movie theaters could be so much better, charge a premium & dominate financially (like http://t.co/SnJsIk8528)...

  2. Convenience: Reserved seating, valet parking, warm embrace of Lyft & Uber including ride-pooling, on-site daycare.

  3. Experience: No commercials before movie; super-comfortable chairs & sofas; food delivery directly to seats; sparkling clean bathrooms.

  4. Food & drink: High-quality food with healthy options; sit-down dining on site; full bars (Lyft & Uber make more practical & safe now).

  5. Variety of screening experiences: Silent, or noisy, or use of phones allowed, or families + kids, or all kids, or dining + movie together.

  6. Use of crowdsourcing & crowdfunding for special screenings; full embrace of group & corporate events; all-you-can-view subscriptions...

  7. I said at the time that I agree with much of what Bill says (https://t.co/Yizp0Zr64F), and I want to expand on the topic further:

  8. New founders in last 10 years have ONLY been in environment where money is always easy to raise at higher valuations. THAT WILL NOT LAST.

  9. When the market turns, and it will turn, we will find out who has been swimming without trunks on: many high burn rate co's will VAPORIZE.

  10. High cash burn rates are dangerous in several ways beyond the obvious increased risk of running out of cash. Important to understand why:

  11. First: High burn rate kills your ability to adapt as you learn & as market changes. Co becomes unwieldy, too big to easily change course.

  12. Second: Hiring people is easy; layoffs are devastating. Hiring for startups is effectively one way street. Again, can't change once stuck.

  13. Third: Your managers get trained and incented ONLY to hire, as answer to every question. Company bloats & becomes badly run at same time.

  14. Fourth: Lots of people, big shiny office, high expense base = Fake we've made it! feeling. Removes pressure to deliver real results.

  15. Fifth: More people multiplies communication overhead exponentially, slows everything down. Company bogs down, becomes bad place to work.

  16. Sixth: Raising new money becomes harder & harder. You have bigger bulldog to feed, need more and more $ at higher and higher valuations.

  17. Therefore you take on escalating risk of a catastrophic down round. High-cash-burn startups almost never survive down rounds. VAPORIZE.

  18. Further, to get into this position, you probably had to raise too much $ at too high valuation before; escalates down round risk further.

  19. Seventh: Even if you CAN raise an up round, you are increasingly likely to incur terrible structural terms like ratchets to chin the bar.

  20. That nice hedge fund investor willing to hit your valuation bar? Imagine him owning 80% of co after down round. How nice will he be then?

  21. Eighth: When market turns, M&A mostly stops. Nobody will want to buy your cash-incinerating startup. There will be no Plan B. VAPORIZE.

  22. Finally, there are exceptions to all this. But if you're reading this, you're almost certainly not one. They are few and far between.

  23. Worry.

In tech, we talk about difference between technical-founder/CEO (product/eng background) vs professional CEO (sales/marketing background).

Written on Aug. 24, 2014, 10:16 p.m.
  1. In tech, we talk about difference between technical-founder/CEO (product/eng background) vs professional CEO (sales/marketing background).

  2. Our general theory is: Easier to teach product innovator how to manage, than it is to teach sales/marketing operator how to innovate.

  3. There are many exceptions in both directions, of course. Mountain is hard to climb either way. Lots of work/learning/adaptation required.

  4. I propose another lens on dynamic: Difference between knowing What & Who, vs knowing How, Where, & When. Bear with me...

  5. Great tech founder/CEOs tend to focus on What & Who: What product to build, and Who to hire/train/retain/motivate to build it.

  6. Great pro CEOs tend to focus on How, Where, & When: How = processes; Where = geographic expansion; When = optimizing business across time.

  7. To succeed at scale, each needs to learn the other skills & hire people who have them: Founder/CEO -> How/Where/When; Pro CEO -> What/Who.

  8. The challenge: Usually easier to hire skilled business professionals who know How/Where/When than What/Who. Fishing from unbalanced pool.

  9. The trap: Only nailing What/Who can carry startup a long way, but only nailing How/Where/When = slow road to zombieland and company death.

  10. Ultimately = team-building for both paths. But dynamic different & differently challenging in each direction; requires open discussion.

  11. Addendum: The Why = the mission. Ideally beyond just the company's success. Increasingly important for all paths.

  12. Addendum: The truly great tech CEOs have mastered all of these: What, Who, How, Where, & When... and Why.

A thing I believe that few believe: Almost all Silicon Valley startup ideas from qualified founders = great ideas. But some are too early.

Written on Aug. 21, 2014, 3:15 p.m.
  1. A thing I believe that few believe: Almost all Silicon Valley startup ideas from qualified founders = great ideas. But some are too early.

  2. Track startups over multiple decades, what you find is that most ideas do end up working. It's much more a question of when not if.

  3. This is interesting for several reasons. First, it means that criticism of the form that will never happen is usually misguided & wrong.

  4. Second, it means that a much bigger risk for founders is too early, vs wrong or too late. Often doesn't match feedback from others.

  5. To quote Peter Thiel, it is often better to be the last company to market (hit timing right & take down the entire market) vs the first.

  6. Third, when you have the timing right, you almost always feel like you're too late. Terrified you've missed the window = great sign.

  7. When idea X has been in the air, with repeated attempts to build X, yet most customers are not yet doing/using X, it's never too late.

  8. Fourth, founders by definition live in the future, see a world that doesn't yet exist & try to make it so. Nailing timing = hardest thing.

  9. Which is often why more pragmatic founders end up building the big & important companies -- the idealists were just too early.

  10. Fifth, therefore, most of the great ideas for the next two decades are already known. In labs, in failed startups, in big co prototypes.

  11. Those ideas are being dismissed now since the early attempts have't worked. This has the opposite predictive value vs what people think.

  12. The key question is: What ideas are widely dismissed today due to having been tried & failed? Answer is the codex to the next 20 years.

I really don't get people who go out of their way to crap on the hard work and efforts of founders and startup teams.

Written on Aug. 20, 2014, 2:18 a.m.
  1. I really don't get people who go out of their way to crap on the hard work and efforts of founders and startup teams.

  2. The simple form of such crapping is pure sour grapes. The advanced version is Silicon Valley is not trying to solve big problems.

  3. In honor of today's outstanding YC demo day, I'll reprise some thoughts from my July 7 tweetstorm on this cynical and pointless canard.

  4. There are six logical problems with the false choice of make trivial apps for 20-something SF hipsters vs do things that matter.

  5. First, make trivial apps vs do things that matter are not actually in conflict-there's plenty of room and plenty of money to do both.

  6. Second, it's often hard to tell which is which up front. Almost all big world-changers were dismissed by critics as trivial at first.

  7. Third, observer bias: Only read consumer tech blogs, only go to consumer tech conferences, think SV only works on consumer tech.

  8. Fourth, battling cynical critiques: Founders who articulate the big vision for changing the world get called arrogant and vainglorious.

  9. Both criticisms leveled with no cognitive dissonance: Founders either not pursuing big ideas, or out of control egomaniacs if they are.

  10. Fifth, subtext often that communication tech/apps in particular somehow aren't important or don't matter, vs energy, education, etc.

  11. Why? Communication is the foundation of collaborative work, which is how all the important problems gets solved. People working together.

  12. Sixth: Anyone who thinks SV can be doing more/better/different, come join us and participate in building new things, products, companies.

  13. The central truth of Silicon Valley is that there's always more to do, and there are always new opportunities to build & contribute.

  14. I couldn't be more proud of today's YC amazing demo day crop, spanning more problem domains than ever. Silicon Valley spirit is thriving.

Tonight I'm tickled pink to be able to talk about our new investment in Buzzfeed! http://t.co/dMA3nKLZL7

Written on Aug. 11, 2014, 1:05 a.m.
  1. Tonight I'm tickled pink to be able to talk about our new investment in Buzzfeed! http://t.co/dMA3nKLZL7

  2. BuzzFeed has technology at its core. Its 100+ person tech team has created world-class systems... Engineers are first class citizens.

  3. And then on top of its technology core, Buzzfeed's reporting team is now routinely committing breathtaking investigative journalism...

  4. We are very excited to work with everyone at Buzzfeed to help them realize their dreams of a profoundly important new media institution.

Something I believe that a lot of people I know believe: We live in Golden Age of journalism as measured by quality of top contributors.

Written on Aug. 10, 2014, 8:38 p.m.
  1. Something I believe that a lot of people I know believe: We live in Golden Age of journalism as measured by quality of top contributors.

  2. I've collected 238 members of the press I most respect into this Twitter list: https://t.co/KddFTkoRla -- It's a joy to read every day!

  3. As I've discussed before http://t.co/hjZsKprQ5i: Great unexpected side effect of Internet = best journalists have far broader reach now.

  4. I can't resist singling out some who fall into category of drop everything I'm doing to read anything they write -- out of many.

  5. Over next 5-10 years I think we'll be able to build more tools/systems for reporters like these to maximize their scope & opportunity.

  6. Quality of work + explosion of journalism entrepreneurship + rise of 5 billion smartphone world + new tools/systems = interesting times!

Outstanding new op-ed by best living economic historian, Joel Mokyr at Northwestern: http://t.co/FjDWMbQygG

Written on Aug. 10, 2014, 11:53 a.m.
  1. Outstanding new op-ed by best living economic historian, Joel Mokyr at Northwestern: http://t.co/FjDWMbQygG

  2. There is nothing like a recession to throw economists into a despondent mood. Much as happened in the late 1930s...

  3. The economic growth experienced through the 20th century, they tell us, was fleeting. Our children will be no richer than we are.

  4. What is wrong with this story? The one-word answer is technology...digital codification of information = reinvention of invention.

  5. [Terms] like 'IT' don't begin to express the scope of the change...[array of] tools that the digital age places at science's disposal.

  6. The consequences are everywhere, from molecular genetics to nanoscience to research in Medieval poetry.

  7. As science solves problems that were not even imagined, inventors, engineers and entrepreneurs are waiting in the wings...

  8. ...to design new gizmos and processes based on the new discoveries that will continue to improve our lives.

  9. The economy may be facing some headwinds, but the technological tailwind is more like a tornado. Fasten your seat belts.

  10. So: If everything is so good, why is everything so bad? Why the gloominess of so many of my colleagues?

  11. [GDP and productivity] work for a steel-and-wheat economy, not [ours]... mismeasure the contributions of innovation to the economy.

  12. Many new things are expensive to design, but copy at low/zero cost: Contribute little to GDP even if consumer welfare impact is large.

  13. I highly recommend Mokyr's books: http://t.co/qM1ymOPSfF http://t.co/vcIg2VxV63 http://t.co/59rALaWK8r

  14. And of course Steven Pinker on the broad perspective of our era: http://t.co/uWIcRQ5LZr http://t.co/4STbPp97de

Something I believe that nobody I know believes: Woodward&Bernstein Watergate coverage precipitated 40yr collapse of trust in print news.

Written on Aug. 10, 2014, 11:13 a.m.
  1. Something I believe that nobody I know believes: Woodward&Bernstein Watergate coverage precipitated 40yr collapse of trust in print news.

  2. That long slow slide of trust can be seen, among other places, in Gallup polls over the years: http://t.co/G4LrmPHGUQ

  3. After Nixon resigned 40 years ago this weekend, Washington Post Watergate coverage became exemplar for entire next generation reporters.

  4. Political press became obsessed with unearthing scandal, which metastasized throughout print journalism. Gunning for Pulitzer bait.

  5. There are clearly scandals that need to be unearthed, like Watergate. BUT: Endless scandal frenzy is exhausting and demoralizing.

  6. Particularly when applied indiscriminately across news landscape, and particularly when extrinsic press motivations are so clear.

  7. Irony is we now know Woodward&Bernstein less reported Watergate than had story fed to them by Mark Felt, partisan in internal FBI battle.

  8. I think the 40 year echo effects of Watergate have more to do with the existential crisis of newspapers than anyone would ever admit.

  9. As news consumers, endless barrage of scandal, tragedy, and conflict has real psychological effects. Makes world seem worse than it is.

  10. Followup reading that provokes thought: http://t.co/KgLWT1mJV7 http://t.co/Br1nlaLbGj http://t.co/XGwZRQp3w7

Lessons learned by managers and shareholders of large regulated financial institutions for the next financial crisis:

Written on Aug. 7, 2014, 5:36 a.m.
  1. Lessons learned by managers and shareholders of large regulated financial institutions for the next financial crisis:

  2. There is no risk of individual executive criminal prosecution whatsoever.

  3. Bailouts are guaranteed, particularly for bondholders, for all but the weakest members of the herd.

  4. The one thing that will get punished is acceding to the government's request/demand for stronger companies to buy weaker companies.

  5. Ultimate fines will be levied against your shareholder base 8 years in the future, not your shareholder base when the sins are committed.

  6. Too big to fail institutions will be allowed to become bigger than ever, increasing their safety buffer for next time.

  7. And for bonus points, regulatory barriers against new competition will be raised, not lowered, further entrenching incumbents.

  8. AND: Too big to jail is real, according to the Attorney General of the United States: http://t.co/pfQqZL1fWu

  9. AND: Regulators on whose watch the last crisis happened, will be allowed to become even bigger and more powerful.

  10. Followup reading, from US Federal Judge Jed Rakoff: http://t.co/FkdqMfYwbn

  11. Followup viewing: Both nonfiction Too Big To Fail film and fiction Margin Call film are excellent at capturing the 2008 crisis.

Really interesting business experiment starting at Procter & Gamble: http://t.co/uHbjdPusAK

Written on Aug. 5, 2014, 12:20 a.m.
  1. Really interesting business experiment starting at Procter & Gamble: http://t.co/uHbjdPusAK

  2. P&G will sell or exit 90-100 mostly minor brands in bold attempt to refocus the business behind its 70-80 remaining best-selling brands.

  3. 'Less will be much more,' P&G CEO told analysts. 'The objective is growth... We're going to be much more agile and adaptable.'

  4. I think a majority of big company CEOs think they should do this in their own companies. But few ever pull the trigger. Too scary.

  5. A common thing you hear at big companies is SKU proliferation -- sheer # of items for sale. Bloats organization & makes action harder.

  6. Steve Jobs legendarily used strategy of cutting brands & SKUs for Apple's turnaround. But few CEOs have followed suit in last 15 years.

  7. Like Steve, AG Lafley at P&G is one of the most respected CEOs in the world. If this works for P&G as well as it did at Apple...

  8. ...I think the odds go way up that many big company CEOs will pull the trigger on the same strategy. Could be transformative for business.

  9. The stakes are high: Whether, and how, big companies will be able to grow their businesses & their # of workers in the future.

  10. Further, whether/how big companies will invest in new product creation in the future. Paring the old can be staging for creating the new.

  11. Of course, Larry Page is busily ignoring Steve Jobs' advice to do the same thing at Google! And Jeff Bezos is furiously expanding Amazon.

  12. No absolutes, but I think it's very healthy for every big company to consider: How to best set up to grow & create new things?

  13. Lyft and Lyft Line are an archetypal example of how Silicon Valley is going straight at the hard problems, in this case transportation.

  14. A growing number of people know about the amazing consumer utility and convenience created by a ride on demand whenever you want.

  15. And in parallel, services like Lyft make it possible for people who may otherwise not be able to make car payments to keep their cars.

  16. In many cities, this results in a triple win: Consumer convenience, driver economic benefits, and improved business/tourism environment.

  17. But beyond that, as Lyft and its peers grow, ride sharing becomes increasingly convenient and affordable as *alternative* to owning a car.

  18. This leads to environmental benefits: Fewer cars needed -> less natural resource utilization; Network efficiency -> fewer miles driven.

  19. Online supply/demand matching eliminates need for cars-for-hire to drive around & look for riders. Network optimization in bits not atoms.

  20. Lyft Line is especially environmentally friendly: Facilitates multiple people riding together on same route, still with high convenience!

  21. Everyone in world wants equivalent to upper-middle-class American lifestyle. Services like Lyft make possible without destroying planet.

  22. Few new tech's deliver so much to so many: riders, drivers, car owners, cities, environment. And to think it just looks like an app :-).

  23. Closing note: Lyft Line is classic peace dividend of smartphone wars -- not possible pre universal smartphones. http://t.co/Lj8FpHXIBj

Answer to last tweetstorm (https://t.co/TkBZV57iI7) is: 1964, in the International Socialist Review -- http://t.co/l9q7kwnl5U

Written on Aug. 3, 2014, 10:07 p.m.
  1. Answer to last tweetstorm (https://t.co/TkBZV57iI7) is: 1964, in the International Socialist Review -- http://t.co/l9q7kwnl5U

  2. Every single argument made by today's technology will eat all the jobs brigade is in there, from 50 years ago.

  3. Including the claim that this time is different, that we finally reached the tipping point where the Luddite Fallacy would come true.

  4. And including the demand for massive additional government intervention in the economy to correct the resulting inherent structural flaw.

  5. Of course, since 1964, enormous numbers of new jobs have been created and quality of life in the US is way up at all income levels.

  6. Identifying the fallacies and flaws in logic in the 1964 manifesto given the 50 years that followed is an interesting exercise...

  7. ...and, I propose, identical to identifying the fallacies and flaws in logic in the equivalent arguments today.

  8. Perhaps most interesting change in era: Advocates then were proud to call themselves Socialists. Less so today. History is speaking to us.

Fun: Here is a manifesto on technology eats all the jobs which I did not write. Guess who wrote it and when: https://t.co/WPVyr2mKj6

Written on Aug. 3, 2014, 9:27 p.m.
  1. Fun: Here is a manifesto on technology eats all the jobs which I did not write. Guess who wrote it and when: https://t.co/WPVyr2mKj6

  2. Computers and the Internet...result in a system of almost unlimited productive capacity which requires progressively less human labor.

  3. Potentially unlimited output can be achieved by systems of machines which will require little cooperation from human beings.

  4. As machines take over production from people, the men and women who are displaced become dependent on minimal government welfare.

  5. The existence of this paradox is denied or ignored by conventional economic analysis.

  6. Capitalist system designed to produce increasing quantity of goods; distribution of purchasing power would occur almost automatically.

  7. Continuance of the income-through-jobs link now acts as main brake on the almost unlimited capacity of a tech-based productive system.

  8. Adequate distribution of goods+services...: Not how to increase production but how to distribute the abundance created by technology.

  9. Underlying cause of unemployment: capability of machines is rising more rapidly than the capacity of many human beings to keep pace.

  10. As a first step to a new consensus it is essential to recognize that the traditional link between jobs and incomes is being broken.

  11. We urge that the government undertake an unqualified commitment to provide every individual with a basic income as a matter of right.

  12. Distribution of abundance in technological society must [have] criteria strikingly different from economic system based on scarcity.

  13. Bonus points: Identify the publication, and its ideological affiliation, in which this manifesto first appeared.

Common thing one hears in US is Political system broken; Founding Fathers never intended politics to be dominated by moneyed interests.

Written on July 13, 2014, 9:02 p.m.
  1. Common thing one hears in US is Political system broken; Founding Fathers never intended politics to be dominated by moneyed interests.

  2. In 1789, George Washington was elected President. Only 6% of the population can vote. http://t.co/PpX4aLQcM8

  3. Not until 1856 was voting expanded even to all white men (eliminating property ownership requirement) -- http://t.co/PpX4aLQcM8

  4. 1868, voting extended to former slaves (at least in theory), but still explicitly defined as male.

  5. Women could first vote starting in 1890 in Wyoming (!). In 1920, 19th Amendment passed, giving women right to vote throughout US.

  6. Throughout late 1800s and early 1900s, lots of battles for and against rights of other groups e.g. Native Americans & immigrants to vote.

  7. Of course, massive battles in the 1960's to extend practical right to vote to African-Americans; some battles continue to this day.

  8. Not until 1971 (year I was born) did voting age get lowered to 18, despite 18-21 year olds being conscripted & sent to war throughout.

  9. Founding Fathers arguably designed US system specifically to be dominated by moneyed interests, aka white male Protestant landowners.

  10. We have far broader-based voting and political participation today than ever before, due to hard work by many activists over 200 years.

  11. And we're still by no means perfect; lots of progress yet to be made. But we're leaps and bounds ahead of 50-100-150-200 years ago.

  12. And no, I'm not watching the World Cup finals :-).

  13. Ralph is a unique combination of business brilliance, outstanding judgment, and human warmth. He makes everyone around him better.

Only hire, and only train/motivate/incent your managers to hire -- don't optimize efficiency, don't do performance management, don't fire.

Written on July 8, 2014, 11:35 a.m.
  1. Only hire, and only train/motivate/incent your managers to hire -- don't optimize efficiency, don't do performance management, don't fire.

  2. Founders, sell too much of your own personal stock too quickly, alienating your employees and questioning your long-term commitment.

  3. Let private stock sales by employees get out of hand: create hit-and-run culture and take on burdens of being public before going public.

  4. Dilute the s*** out of cap table: be sloppy & undisciplined w/stock grants to early employees, plant morale land mine for later employees.

  5. Maximize absolute valuation of each growth round: make later rounds harder and harder to achieve, until you trigger disastrous down round.

  6. Let non-SV investors suck you into terrible structural terms on growth rounds: guarantee massive trauma if anything goes slightly wrong.

  7. Go public too soon, before you're a fortress, before you can withstand all the assaults: ending in stock price death spiral & train wreck.

  8. Pour huge money into overly glorious new headquarters, signaling to employees we've made it, we're amazing, then repeat two years later.

  9. Confuse conference circuit & party scene with actual work. Encourage alcohol & drugs, party culture in company, value ballers over nerds.

  10. Refuse to take HR seriously: allow terrible internal manager & employee behavior to catalyze into catastrophic ethical & legal crisis.

  11. And the one that will actually kill you: Assume more cash is always available at higher & higher valuations, forever.

One persistent canard from would-be SV critics is Silicon Valley isn't building/funding the right things, aka solutions to big problems.

Written on July 7, 2014, 11:46 a.m.
  1. One persistent canard from would-be SV critics is Silicon Valley isn't building/funding the right things, aka solutions to big problems.

  2. There are six logical problems with the false choice of make trivial apps for 20-something SF hipsters vs do things that matter.

  3. First, make trivial apps vs do things that matter are not actually in conflict-there's plenty of room and plenty of money to do both.

  4. Second, it's often hard to tell which is which up front. Almost all big world-changers were dismissed by critics as trivial at first.

  5. Third, observer bias: Only read consumer tech blogs, only go to consumer tech conferences, think SV only works on consumer tech.

  6. Founders of non-consumer-tech startups routinely find same pundits mounting criticism have little interest in hearing about other domains.

  7. This is exacerbated by SF-centric consumer tech party scene--other domains in SV don't have the same party culture, just nerds at work.

  8. New arrivals to SV get sucked into SF party scene, and never make it to the South Bay industrial parks where everything else is happening.

  9. Fourth, battling cynical critiques: Founders who articulate the big vision for changing the world get called arrogant and vainglorious.

  10. Both criticisms leveled with no cognitive dissonance: Founders either not pursuing big ideas, or out of control egomaniacs if they are.

  11. Fifth, subtext often that communication tech/apps in particular somehow aren't important or don't matter, vs energy, education, etc.

  12. I think this is 100% incorrect: Communication tech/apps including Internet are the foundation for everything else we'll do for 100 years.

  13. Why? Communication is the foundation of collaborative work, which is how all the important problems gets solved. People working together.

  14. Sixth: Anyone who thinks SV can be doing more/better/different, come join us and participate in building new things, products, companies!

  15. Jump in, the water's warm! SV draws talent from all over the world & all walks of life; nothing preventing any critic from contributing.

  16. As my old boss Jim Barksdale used to say, We have plenty of uniforms your size. Many opportunities to contribute & make a difference!

  17. And, of course, tech startup ecosystem now expanding worldwide. Opportunities to contribute from anywhere abound, linked via Internet.

  18. Ten + one ways to grievously damage your high-growth tech startup, and Silicon Valley in the process:

Halt And Catch Fire takes flak from critics for being melodramatic--yet captures real emotional intensity of actual tech startups well.

Written on July 7, 2014, 6:16 a.m.
  1. Halt And Catch Fire takes flak from critics for being melodramatic--yet captures real emotional intensity of actual tech startups well.

  2. Joe, Cameron, Gordon, Bos all accurate archetypes. I know dozens of each. Must come together to do great things. Can't alone.

  3. Tonight's episode best since pilot -- without spoiling, key is both Cameron and Gordon are right. Real tension at heart of many startups.

  4. Again without spoiling, final question asked by Cameron's new program still at the heart of our industry 30 years later.

  5. Show also shows why founders often don't start 2nd company, or sometimes even talk to former partners after company is over. Too intense.

  6. Finally, show is dead on about overwhelming market power of IBM at that time. Seeds of collapse had already been planted but no one knew.

Cleaned-up tweetstorm on the reverse rules of new technology -- expressed ironically! -- and labeled as such :-).

Written on July 6, 2014, 8:18 p.m.
  1. Cleaned-up tweetstorm on the reverse rules of new technology -- expressed ironically! -- and labeled as such :-).

  2. One thing we know about new gadgets: They can be judged based on their first versions since that's what they'll look like forever. #irony

  3. Each new technology arrives into the world w/the perfect universal use case pre-identified. If it doesn't, it flops & dies forever. #irony

  4. Any new technology needs immediate widespread cultural acceptance. If it's just used by nerds early on, it'll never go mainstream. #irony

  5. New technology products must be immediately cash-flow profitable; otherwise you have clear evidence idea was stupid from the start. #irony

  6. Any new tech product that isn't immediately affordable for everyone will never decline in price & will just exacerbate inequality. #irony

Gallup surveys on confidence in institutions are endlessly interesting: http://t.co/qHx0Ce6YCz

Written on July 1, 2014, 1:28 a.m.
  1. Gallup surveys on confidence in institutions are endlessly interesting: http://t.co/qHx0Ce6YCz

  2. High-trust US institutions (above or near 50%): Military, small business, police, churches.

  3. Low-trust US institutions: Medical system, Supreme Court, Presidency, public schools, banks.

  4. Almost-no-trust US institutions (<25%): Justice system, newspapers, organized labor, big business, Internet news, TV news, Congress.

  5. Conventional glass-half-empty argument is terrible erosion of trust & competency in key institutions, decline & fall of the United States.

  6. Alternate glass-half-full argument: Large centralized institutions *shouldn't* be highly trusted: monopolies/oligopolies let people down.

  7. Alternate arg cont'd: Lack of trust in institutions motivates construction of competitive alternatives better attuned to we the people.

  8. Alternate arg cont'd: Large centralized institutions were 20th century; flexible, responsive, accountable organizations for 21st century.

  9. I don't think the answers are at all straightforward. But I think this debate is going to inform a lot of next 30 years of politics.

  10. Will be interesting to see how many large centralized institutions can win those arguments starting with such low confidence levels.

New study on interplay between diversity and success in VC investing in startups -- really interesting -- http://t.co/Tj4WBKK9YQ

Written on June 27, 2014, 5:48 p.m.
  1. New study on interplay between diversity and success in VC investing in startups -- really interesting -- http://t.co/Tj4WBKK9YQ

  2. The more [ethnic and educational] affinity there is between two VCs investing in a firm, the less likely the firm will succeed. !!

  3. VCs have strong tendency to team w/other VCs whose ethnic & educational backgrounds are similar... turns out to be bad for business.

  4. Two VCs from same undergraduate school 34.4 percent more likely to collaborate...and increased by 39.2 percent if same ethnicity.

  5. Odds of success of invested company down 17% if two VCs worked at same company; -19% same undergrad school; -20% same ethnicity.

  6. The lack of success among similar investors seemed to lie in the decisions that *followed* the investment...due to 'groupthink'.

  7. Side note: Last names of the three researchers who wrote the paper -- Gompers, Xuan and Mukharlyamov :-).

Late night tweetstorm addendum to earlier series on unbundling (https://t.co/ecgBveoiMj) and rebundling (https://t.co/0fw9alDKh3)...

Written on June 25, 2014, 10:03 a.m.
  1. Late night tweetstorm addendum to earlier series on unbundling (https://t.co/ecgBveoiMj) and rebundling (https://t.co/0fw9alDKh3)...

  2. Younger followers asked me to expand on DEC/Sun/Microsoft unbundling: older history, therefore useful to study as precedent for our time.

  3. Once upon a time (1950s-1960s), businesses mostly bought computers from IBM, & IBM also bundled in all the software & services you'd need.

  4. IBM bundled offerings were extremely expensive, while demand for computers spread to departments & smaller businesses who had less $.

  5. So DEC unbundled the computer itself from the total bundle & sold it for less $ to customers who could then write their own software.

  6. But a DEC computer itself was a bundle of proprietary components: VAX hardware, VMS operating system, RDB database, etc.

  7. In the 80's, Sun came along & sold a similar power computer to DEC's for 4 the cost w/unbundled components: Unix OS + Motorola chips.

  8. DEC responded by going upmarket & becoming more like IBM; Sun & other Unix workstation vendors chewed through their market + grew it more.

  9. Critical point: At point when DEC started to tip over, its integrated offering was as close to perfect as our industry will ever see.

  10. Ask people who were working then (like me) and they'll tell you to this day that they miss VAX/VMS & how productive they could be on it.

  11. Meanwhile Microsoft & Intel came along and fully implemented the unbundled low-cost computer, first PCs and then Windows-based servers.

  12. Sun responded to Microsoft/Intel by rebundling: Rebuilding itself in DEC's image with Sparc chip, Solaris OS -- proprietary HW+SW stack.

  13. Again, Sun got close to perfection as an integrated system when unbundled Microsoft+Intel & Linux+Intel went broad & forced sale of co.

  14. Picking up thread with Microsoft: Microsoft has spent last 15 years rebundling, rebuilding itself in DEC's image much like Sun. Now + HW!

  15. Once again, bundled Microsoft stack very close to perfect just as it comes under attack from next unbundling wave: mobile + cloud.

  16. Fascinating twist: Under new leadership, Microsoft 2014 now committed to unbundling -- full app + cloud support for iOS, Android, Linux!

  17. Microsoft taking new unbundling strategy offensively vs new titans who are bundling as fast as possible! Apple, Google, Amazon, Oracle.

  18. The point? I don't know much about what our industry will look like in 50 years, but I'm quite confident these same dynamics will apply.

The flip side of unbundling: Later on, the unbundlers tend to try to rebundle in the image of whatever they unbundled.

Written on June 24, 2014, 11:06 p.m.
  1. The flip side of unbundling: Later on, the unbundlers tend to try to rebundle in the image of whatever they unbundled.

  2. So Yahoo adds an ISP (https://t.co/JrlrVoY9Vo), and Google adds email/IM/sports-scores/stock-quotes.

  3. Twitter changes its user profile page to look more like Facebook :-).

  4. Sun unbundled DEC with commodity components, then re-bundled into a proprietary computing stack just like DEC w/Solaris, Sparc, etc.

  5. Microsoft likewise unbundled DEC minicomputers w/PC OS + tools, then rebundled into DEC-like integrated stack now including hardware (!).

  6. Paraphrasing Harvey Dent: You either die a hero or you live long enough to see yourself become the company you first competed with.

  7. And then sometimes the rebundlers realize what they're doing and try to reverse course. E.g. Microsoft building apps for iOS & Android.

  8. And thus the cycle of life repeats with yet more unbundling :-).

A story of unbundling in the tech industry: 20 years of consumer Internet evolution -- http://t.co/qUCWjc4wkF

Written on June 24, 2014, 9:47 p.m.
  1. A story of unbundling in the tech industry: 20 years of consumer Internet evolution -- http://t.co/qUCWjc4wkF

  2. One upon a time there was AOL, which was a completely integrated Internet access/information/communication service.

  3. Then Yahoo came along and unbundled the information/communication parts like email/IM/sports-scores/stock-quotes from the access service.

  4. One of the things you could do on Yahoo was search, then Google came along and unbundled that.

  5. You can search for anything on Google, including people; Facebook came along with a much better way to just search for people.

  6. Three things you can do on Facebook are messaging, photo sharing, and status updates; therefore Whatsapp, Instagram, and Twitter.

  7. And yes, Yo unbundles the creation & existence of a message from the contents of a message, unbundling Whatsapp and Twitter :-).

  8. The part people often miss is that you can get extremely powerful second/third order effects at each step with his pattern.

  9. The entrepreneurs generally have a pretty good sense of this when they're doing it, but it doesn't become clear to others until later.

  10. This is a pattern what we love to fund: unbundle X from Y, but then use the liberation of X as leverage to do amazing new things with X.

  11. And the howls of press and analyst outrage at the apparent stupidity of each unbundling are very helpful for keeping valuations down :-).

1950's + 1960's + 1970's + 1980's + early 1990's view of computer technology = Nerds!

Written on June 24, 2014, 4:43 a.m.
  1. 1950's + 1960's + 1970's + 1980's + early 1990's view of computer technology = Nerds!

  2. Late 1990's view of computer technology = Everyone will get rich!

  3. 2001's view of computer technology = The nerds screwed us!

  4. 2003's view of computer technology = We knew those nerds were wrong all along!

  5. 2009's view of computer technology = Those nerds are completely out of ideas!

  6. 2013's view of computer technology = Those nerds and all their crazy ideas are going to destroy all the jobs!

  7. And now = Those nerds are completely out of ideas again, and now they're having sex too! http://t.co/Du7pMU6seU http://t.co/GF5L2qXwj3

  8. Stay tuned for more updates from our favorite show, As The Nerds Turn :-).

Today we're thrilled to be able to talk about newest investment--Tanium. We're investing $90M; first VC investment since co founded 2007.

Written on June 22, 2014, 11:04 p.m.
  1. Today we're thrilled to be able to talk about newest investment--Tanium. We're investing $90M; first VC investment since co founded 2007.

  2. Tanium --> people responsible for large networks of computers & software, what Google --> people on the Internet. Don't say that lightly.

  3. Company's operating under radar, but its customers could not be more enthusiastic. Product must be seen to be believed. It's amazing.

  4. Tanium is real-time query *and* control--100s of Ks of computers & virtual machines--with natural language queries. http://t.co/q4uYZl4qaX

  5. We are thrilled to be working with genius founders David & Orion Hindawi and their colleagues. Masters of the art & science of computing.

  6. More info: http://t.co/q4uYZl4qaX http://t.co/Ro5qnQPwPU http://t.co/uZrYn0Y7MT or email info@tanium.com :-)

  7. Now Wink will provide a standard software layer across all of Quirky's products plus many others including GE, Honeywell, and Philips.

  8. Quirky partner Home Depot now sells 600 smart-home products, six times as many as it did two years ago. The revolution is happening now.

  9. Important to note this isn't just about toys for rich people; it's about energy conservation, water conservation, security, and safety.

The thing I find most interesting about Marx is how much he hated the poor and dispossessed: http://t.co/A3Y9U4DsEB

Written on June 20, 2014, 8:35 p.m.
  1. The thing I find most interesting about Marx is how much he hated the poor and dispossessed: http://t.co/A3Y9U4DsEB

  2. Lumpenproletariat: Layer of working class lost to useful production, of no use to the struggle, an impediment to the classless society.

  3. Translated as slum workers or the mob...class of outcast, degenerated and submerged elements within population of industrial centers.

  4. Beggars, prostitutes, gangsters, racketeers, swindlers, petty criminals, tramps, chronic unemployed or unemployables...

  5. Declassed, degraded or degenerated elements...innumerable young people also, who cannot find an opportunity to [become] producers.

  6. Vagabonds, discharged soldiers, discharged jailbirds, escaped slaves, swindlers, mountebanks, pickpockets, tricksters, gamblers...

  7. Maquereaux [pimps], brothel keepers, porters, literati, organ grinders, ragpickers, knife grinders, tinkers, beggars...

  8. In short, the whole indefinite, disintegrated mass, thrown hither and thither. The man did have a way with words.

  9. References: http://t.co/A3Y9U4DsEB, http://t.co/YB1t3fbjSV, http://t.co/eQ4LD3NE1q

  10. $200 a month, teach anyone basic programming skills for entry-level position at AT&T as a data analyst, iOS applications designer, etc.

  11. Harnessing the web to provide effective schooling to the many young Americans for whom college has become a distant, unaffordable dream.

  12. 'We are trying to widen the pipeline,' said AT&T's Charlene Lake. 'This is designed by business for specific skills needed in business.'

  13. Education still offers children from disadvantaged families their best chance at climbing the ladder of success.

  14. One reason for enormous payoff from college degree (2x payoff vs 1979) is that too few young Americans ever earn one.

  15. Peter Lubbers, who runs MOOC developer training for Google says, 'We want all the techniques we know about to get out to the market.'

  16. The 'NanoDegree' offers a set of skills that can be clearly applied to a job--a chunk of knowledge & immediate motivation to acquire it.

  17. Could offer a plausible path to young men/women who may not have time, money or skill to make it through 4-year or even 2-year degree.

  18. US students in bottom quarter of income distribution: college graduation rates only 9%, from 5% 20 years ago. Not *nearly* high enough.

A personal meditation on how we learn, and how how we learn is changing radically for the better due to new technology...

Written on June 20, 2014, 6:56 p.m.
  1. A personal meditation on how we learn, and how how we learn is changing radically for the better due to new technology...

  2. 1989: I go to college & want to learn about how real computers work--VMS & Unix--foundation of all today's phones/tablets/PCs/servers.

  3. Only way: Get a job at a computer lab that owned such a computer; get access to the literal wall of paper manuals-- http://t.co/sGgCIV3jMn

  4. Read those cover to cover; then what? Scientific/technical journals. Only with access to research university library. Until...

  5. 1990: Co-op job at IBM; discovered mainframe (!) search engine of science/tech journals; in 3 days, get printed copies via office mail!

  6. Posed major dilemma: How many papers could the intern request before Big Blue would fire his ***? Tested it into the hundreds.

  7. All other relevant info was on the Internet--IETF RFC's, free software code--but who had access to the Internet in 1990? Not many.

  8. 25 years later: Every smartphone is equiv of million-$ Unix supercomputer from that time, plus *all* that info is freely available online.

  9. Science/tech journals still not completely free, but even those walls falling fast now due to Google Scholar/PDF searches/policy reform.

  10. Re journals: Aaron Swartz doc is out; powerful memory of freedom of information hero & warrior for progress. http://t.co/Vh9G9FuBeZ

  11. Why does this matter? Info is foundation of progress; info about computers is basis of key tool for progress as software eats world.

  12. Means of production being put in hands of the masses for first time in human history. I'm broken record on this, but: profound change.

I am super-proud of my friends at Google and their new push to help get more girls into coding: The things you love are #MadewithCode!

Written on June 20, 2014, 4:21 a.m.
  1. I am super-proud of my friends at Google and their new push to help get more girls into coding: The things you love are #MadewithCode!

  2. Along with several partners, Google is launching Made with Code, an initiative to inspire girls to code. http://t.co/ewOIysOzW0

  3. Including commitment of $50 million to support programs that can help get more females into computer science. !! http://t.co/ewOIysOzW0

One of the special things about our industry is how intellectually generous many of the leading participants are (no, I don't mean me :-).

Written on June 19, 2014, 10:33 p.m.
  1. One of the special things about our industry is how intellectually generous many of the leading participants are (no, I don't mean me :-).

  2. When I arrived in Silicon Valley in Jan 1994, I sought out all of the written material I could on startups & venture capital.

  3. I found exactly two books. An excellent but dry financial analysis of startup returns, and an excellent but dated book by Gordon Bell.

  4. So then I looked for magazines, and found exactly one: Red Herring. Which for several years was the best magazine about startups.

  5. But, in 1994, Red Herring was ~8 (?) memographed pages, cost $12 (?), published every 2 months (?), & available at only a few newsstands.

  6. That was it. I knew there was more material at Stanford & Harvard business schools but I couldn't get to it. There was nothing else.

  7. Today, 20 years later, the difference is *profound*. Many of the leading theorists & practitioners share *huge* amounts of info for free.

  8. That's a big difference in SV, but what I hear every day from people all over the world=what a big difference it's making everywhere else.

  9. A 14-year-old kid in Indonesia w/smartphone has access to 10,000x more info on tech & startups today than I did in Palo Alto 20 years ago.

  10. And the cycle is closing: there is startlingly profound new thinking happening all over the world & coming right back to Silicon Valley.

  11. In our industry, it's hard to underestimate the consequences of a positive feedback loop -- and this is a positive feedback loop.

  12. Assumption *must* be: Tech entrepreneurship all over the world is going to expand a thousand fold in the next 20 years. How could it not?

Lots of Mirth over Yo today (http://t.co/vYHSgl9OOh) but actually there's a fascinating aspect lots of people are missing...

Written on June 19, 2014, 7:06 a.m.
  1. Lots of Mirth over Yo today (http://t.co/vYHSgl9OOh) but actually there's a fascinating aspect lots of people are missing...

  2. Yo is an instance of one-bit communication -- a message with no content other than the fact that it exists. Yes or no. Yo or no yo.

  3. Other instances of one-bit communication: Police siren, flashing stop light, Open sign, light turned on, taxicab roof indicator lit.

  4. But the most interesting instance of one-bit communication is the global missed call phenomenon: http://t.co/EVVISRA40Q

  5. Missed call on mobile phone is used as one-bit comm: Used in South Asia/Philippines/Africa to communicate pre-agreed messages for free.

  6. Aided by fact that missed calls cost nothing to send/receive. In Bangladesh, missed calls are 70% of mobile traffic at any given time.

  7. So the hilarity around Yo includes two problematic biases: Bias that one-bit comm isn't useful, and bias that all markets are like the US.

  8. I'm not saying Yo will be the next $100B social media powerhouse. But instant dismissal makes little sense; let's learn & keep minds open.

Proposed: Unless you're with your kids, what's on screen in your hand = likely more interesting & important than what's around you.

Written on June 19, 2014, 5:21 a.m.
  1. Proposed: Unless you're with your kids, what's on screen in your hand = likely more interesting & important than what's around you.

  2. Sounds extreme, yet how could it not be true most of the time? Screen in your hand contains the entire world vs immediate locality.

  3. This is personal for me: I grew up in very small rural town; Kids in rural towns today grow up way more connected to the world than I did.

  4. People around you not interested in same things? No problem, screen in your hand connects you to like-minded people everywhere.

  5. People around you can't teach you the things you want to know? No problem, screen in your hand gives you all the information you want.

  6. The place where you live doesn't have economic opportunity for you? Screen in your hand gives you access to global opportunities, markets.

  7. Bastiat's seed and unseen: What we see are people staring at their phones. What we don't see are their interactions with the world.

  8. And then it comes back around: Our online virtual world enhances and improves our local physical world & our friends and family in it.

  9. Recommend reading David Gerlenter's Mirror Worlds, on how our virtual & physical worlds improve one another: http://t.co/MXph8TMhvo

Speaking of disruption: Tech people like me can sometimes come across as presumptuous/arrogant re disruption of other peoples' industries.

Written on June 18, 2014, 4:15 p.m.
  1. Speaking of disruption: Tech people like me can sometimes come across as presumptuous/arrogant re disruption of other peoples' industries.

  2. From this side of the aisle, though, it's less smugness, more the result of hard experience and learning from our own lives and careers.

  3. In tech, our *own* businesses are disrupted by technology change and new competitive entrants at whiplash-inducing rates.

  4. It's shocking how quickly you can go from the hot disruptive upstart to the stodgy disrupted incumbent in tech--frequently within 5 years.

  5. I've probably been on the receiving end of disruption 30 times in the last 20 years--almost as many times as I've been on the giving end.

  6. Now, on the one hand, you might say, How can people live like that? ... what's wrong with a little stability?

  7. But, what we see is: Frequent disruption is the handmaiden of rapid progress--and it's a blast to create and work amid rapid progress.

  8. It's not just rapid progress of tech. It's also rapid grown of companies, and even better, rapid development of *people* & their talents.

  9. It's hard to stay in tech for any period of time and not get good at rapid adaptation, skill acquisition... and new product creation.

  10. As software eats the world: Same disruption dynamics always present in tech now applying to many more industries, fields, professions.

  11. Rather than superiority/contempt, what a lot of us feel is deep sympathy/understanding-- even if that's not always how it comes across!

  12. And now we all have the opportunity to learn together--to make many parts of industry/life more innovative/dynamic, better for everyone.

Fascinating question around Christensen's theory of disruption: OK, smart guy, why haven't Apple iPhone/iPad been disrupted by Android?

Written on June 18, 2014, 1:31 a.m.
  1. Fascinating question around Christensen's theory of disruption: OK, smart guy, why haven't Apple iPhone/iPad been disrupted by Android?

  2. I count five possible answers; there may be more...

  3. A. Theory of disruption is hucksterish management consultant fraud. (This one I do not agree with.)

  4. B. Right now Apple *is* the disruptor--iPhones/iPads vs Windows PCs. Many surprised by rapid rise of direct substitution, including me.

  5. C. Apple *is* getting disrupted right now: Android phones outselling iPhones somewhere between 5:1 and 10:1 worldwide right now.

  6. D: Related to C, Apple isn't getting disrupted *yet*, but will be soon. This is what many in Silicon Valley believe, but Apple does not.

  7. E: The most interesting one: Current theory of disruption is incomplete; does not have a broad enough concept of end-user quality.

  8. In this line of argument, disruption theory was born in Microsoft/Intel era, when everyone expected computers to have, um, certain issues.

  9. Apple brilliantly redefined conception what was possible from end-user quality and integration standpoint, against prevailing assumptions.

  10. We have attempted to generalize *this* concept into full stack thinking, which many of today's best startups are also pursuing.

  11. It's possible disruption theory needs to be evolved to accommodate these newer patterns and learnings...

  12. But it's also possible all such full stack patterns are just integrated approaches that themselves will be disrupted in the future.

  13. Time will tell. In Silicon Valley, these topics are central and being debated both in actions and words by ultra-smart people every day.

A few thoughts on timing and staging of capital into modern tech startups -- start with fact that 2003-20011 seed rounds were ~$500K-1M.

Written on June 14, 2014, 4:09 a.m.
  1. A few thoughts on timing and staging of capital into modern tech startups -- start with fact that 2003-20011 seed rounds were ~$500K-1M.

  2. So then think about the startup that's raised $3-4M or even $5-6M in seed funding that goes to raise a Series A from VC firms.

  3. VC looks back across the table: You're not raising a Series A, you're raising a Series B--you already raised your Series A [in seed $].

  4. This can take the startup by surprise, because it really affects how VCs think about progress and milestones, key to raising new round.

  5. VCs assume Series A is to build product + get first beta customers; Series B is to build the business around the product + get to revenue.

  6. So a startup that raised as much $ as a Series A in seed funds, but hasn't achieved actual Series A milestones, can be in real trouble.

  7. VC says: You said you're raising A but you're actually raising B, and you haven't accomplished enough to merit a B. Thank you, but pass.

  8. So the risk of calling $3-4-5-6M seed raises seed is that founder can fool himself/herself heading into the first real VC raise.

  9. The rise of the New Series A by firms like A Capital is intended to address this issue head on. http://t.co/85ry6XtiiE

  10. In effect, a $3-5M seed round or a $3-5M New Series A is a recreation of the original conception of an A round from historical VC.

  11. Takeaway for founders? Think very hard about timing and staging of capital vs progress and milestones; matters a lot for raising A/B/C.

The VA scheduling system scheduled its first appointment in April of 1985. It has not changed in any appreciable manner since that date.

Written on June 11, 2014, 7:51 p.m.
  1. The VA scheduling system scheduled its first appointment in April of 1985. It has not changed in any appreciable manner since that date.

  2. Said Philip Matkovsky, assistant deputy under-secretary of the VA. http://t.co/558HmaTfq7

  3. So, watch Halt and Catch Fire, while thinking to yourself, current VA scheduling system was being built at the same time.

  4. And then think about the odds of any private company still being in business with a core system running unchanged for 30 years.

On AI & the Turing Test: Recently the Internet got all fired up--did a software bot pass/not pass the Turing Test: http://t.co/X4MRbdhjfk

Written on June 10, 2014, 2:51 p.m.
  1. On AI & the Turing Test: Recently the Internet got all fired up--did a software bot pass/not pass the Turing Test: http://t.co/X4MRbdhjfk

  2. Turing Test, proposed by uber-genius Alan Turing in 1950, is: Can a software bot convince a human it's also human, via text chat?

  3. My view is that Turing Test has always been malformed, humans are too easy to trick, passing test says almost nothing about software.

  4. But Marc, Alan Turing was the genius of all time, and you're just some dude on Twitter. What the hell, man? That's a good point!

  5. Turing said something else that I think is far more relevant, which he announced loudly in the executive cafeteria at Bell Labs in 1942:

  6. And I think that's actually what happened, and it happened in the form of enterprise software: the code that runs businesses today.

  7. Yep, that's right: Actual AI is SAP, Peoplesoft, Oracle, http://t.co/IQqvVD9Xj7, Workday, Netsuite, Great Plains, and a thousand others.

  8. AI turned out to be the last thing anyone expected. Banal.

  9. Corollary: If you're not worried about enterprise software taking all the jobs, then you shouldn't worry about AI taking all the jobs.

How important is new communication technology both to maintenance of state control & subversion of state control? http://t.co/mpxdNs0dzf

Written on June 9, 2014, 2:37 p.m.
  1. How important is new communication technology both to maintenance of state control & subversion of state control? http://t.co/mpxdNs0dzf

  2. 1989: 40 years into the era of xerography, Soviet Union is moving its photocopiers from guarded, double-locked, steel-covered doors...

  3. ...admitting photocopiers are standard office equipment and not really the grave threat to state security as they were once perceived.

  4. Main Directorate of Protecting Public Order said recent spread to more than 60,000 different Soviet organizations of photocopiers...

  5. ...had made its task of supervising their operation virtually impossible...freed in major step towards glasnost, or political openness.

  6. Inspectors from the Interior Ministry will no longer be checking each machine to ensure that it is locked in secure quarters...

  7. ...that complete registers are kept of the documents copied and that the numbers of copies correspond to tamper-proof counters.

  8. People had faced very long lines [to make copies], as they and their papers were scrutinized at authorized photocopier offices.

  9. Izvestia said 'the primary goal was to prevent the proliferation of undesirable texts' in the Soviet Union...

  10. 'From the first days of Soviet power, even typewriters were put under tight supervision, and that was abolished only recently.'

  11. Chief of Second Department of Main Directorate of Protecting Public Order, said new technology has overtaken controls on photocopiers.

  12. With computers/optical scanners/laser printers 'to produce a whole newspaper'...regulations governing copiers have become pointless.

Have to post this again because it's just too good not to -- who's coming to Silicon Valley from @BW:

Written on June 8, 2014, 9:54 p.m.
  1. Have to post this again because it's just too good not to -- who's coming to Silicon Valley from @BW:

  2. One third of Silicon Valley startups are founded by Indian-Americans.

  3. As of 2010, Asian-Americans are the majority of Silicon Valley tech workforce: 50% vs 40% for Caucasians.

  4. Possibly eye-opening sources of talent in quantity: Japan, Middle East, Vietnam, France, Pacific Islands, Africa, Caribbean.

  5. Silicon Valley is a powerful successful example of the melting pot theory: ignore origin and ethnicity, come together to do big things.

  6. Why we must continue to push to expand access/openness/inclusion to all origins/ethnicities/genders/religions: huge opportunities ahead.

Part 4 of tech gives people superpowers tweetstream (part 3: https://t.co/2qAzQwUY1k) -- implications of the growth of tech superpowers.

Written on June 8, 2014, 8:07 p.m.
  1. Part 4 of tech gives people superpowers tweetstream (part 3: https://t.co/2qAzQwUY1k) -- implications of the growth of tech superpowers.

  2. First, some tech superpowers help us as consumers, but many upgrade us as creators, builders, inventors, designers, artists: *producers*.

  3. While it is true that tech superpowers can replace need for prior manual labor, it's also true that they enhance our ability to produce.

  4. Both sides of the equation are critical, otherwise one collapses into the Luddite fallacy and holds back progress that benefits everyone.

  5. This is also why I think we need an ongoing vigorous social safety net, to help people bridge gaps in their lives as tech leaps forward.

  6. Second, tech superpowers at this point are being applied more or less equally to individuals, governments, and businesses.

  7. In Orwell's vision, government got all the tech superpowers. In our world, everyone seems to be getting them all at the same time.

  8. Same tech advances that enable NSA surveillance also enabled Edward Snowden to walk out of NSA with 1.7 million classified documents.

  9. (Ability to use a web crawler and a thumb drive would have been the fantasy of all time for Kim Philby. Times have changed.)

  10. The impact of tech superpower upgrading of individuals, businesses, and governments radically shifts power balances between the three.

  11. The politics of the next 30 years will be in large part defined by the shifting impact of tech superpowers across indiv, bus, and gov't.

  12. Prior conflicts like Snowden, SOPA, the Arab Spring, and cybercrime are only open gambits -- the real drama is yet to come.

Progressive and smart economist Jared Bernstein on the productivity puzzle of robots eating all the jobs (or not): http://t.co/8Gd2AQugYn

Written on June 8, 2014, 9:15 a.m.
  1. Progressive and smart economist Jared Bernstein on the productivity puzzle of robots eating all the jobs (or not): http://t.co/8Gd2AQugYn

  2. Productivity growth was up 1% last year and has averaged 0.8% since 2011...[a] smooth trend through the numbers.

  3. The trend suggests that the pace of productivity growth has decelerated since the first half of the 2000s: begs an important question.

  4. But the robots-are-coming advocates need to explain why a phenomenon that should be associated with accelerating productivity...

  5. ... is allegedly occurring over a fairly protracted period where the [productivity] trend in output per hour is going the other way.

  6. My own take: We're still coming out of a severe macroeconomic down cycle, credit crisis, deleveraging, liquidity trap.

  7. Prevailing pessimistic economic theories--death of innovation, robots eating all the jobs, crisis of inequality--will fade with recovery.

  8. For bonus points, identify the other tech-driven economic force that could explain low productivity at a time of great tech advancement.

  9. My nomination: Tech-driven price deflation; lowers prices, reduces measured GDP & productivity, while boosting consumer welfare.

Part 3 of tech gives people superpowers tweetstream (part 2: https://t.co/WaEQoefuqz) -- new superpowers coming in the next 10 years.

Written on June 7, 2014, 8:57 p.m.
  1. Part 3 of tech gives people superpowers tweetstream (part 2: https://t.co/WaEQoefuqz) -- new superpowers coming in the next 10 years.

  2. This is a difficult list to make; there are hundreds if not thousands of candidate superpowers on deck that could hit big next 10 years.

  3. Crowdfunding = superpower to instantly raise money from global customers and investors for millions of new ideas, products, businesses

  4. Quantified self = superpower to understand one's own body continuously in real time, optimize health & wellness with high precision

  5. 3D printing = superpower to design and instantly manufacture physical objects with zero inventory, supply chain, or transportation

  6. Combine modern bio, 3D printing, & computing-->prosthetics & exoskeletons; superpower: paralyzed to walk, disabled to abled, bilnd to see

  7. Bitcoin & blockchain = superpower to instantly & safely transact, do business, engage in commerce & trade with everyone in the world

  8. Virtual reality = superpower to experience other places and times, real and created, serious and fun; telepresence replaces transportation

  9. Augmented reality = superpower to know a million times more about the world around you and everything in it; step function increase in IQ

  10. Self-driving cars = superpower to dramatically reduce # of cars on road, turn parking lots into parks, while saving millions of lives

  11. Drones = superpower to easily function in the air just like we do on the ground; see from the sky, deliver in the air, and fly in VR

  12. And yes, robots = superpower to augment human physical capability; easily do physical work too difficult or dangerous for unaided humans

  13. The nature of each of these is that we don't yet live in a world where they are ubiquitous: very difficult to imagine true implications.

  14. We therefore can predict only a very small fraction of the things that people will do with these superpowers once they have them.

Part two of tech gives people superpowers tweetstream (part 1: https://t.co/2Bx6rZwefV) -- new superpowers just in the last ~10 years:

Written on June 6, 2014, 3:50 a.m.
  1. Part two of tech gives people superpowers tweetstream (part 1: https://t.co/2Bx6rZwefV) -- new superpowers just in the last ~10 years:

  2. Google, Wikipedia = superpower to ask any question, get any answer, as many as you want, for free -- from the world's total knowledge base

  3. Facebook = superpower to always be connected with everyone important in your life, regardless of geography, all the time, for free

  4. eBay, Etsy, Alibaba = superpower to take goods you make or want to resell to a global market of buyers, with transparent & fair pricing

  5. GPS + Google Maps + smartphones = superpower to never be lost, always be able to find anything, and always know where your kids are

  6. Spotify, Beats = superpower to listen to the entirety of recorded music in human culture as much as you want, anytime you want

  7. Lyft, Uber, AirBNB, HotelTonight = superpower to be able to easily move around and then stay places with transparency and safety

  8. Skype, Slack, Asana = superpower to form into teams and collaborate on projects with people all over the world regardless of geography

  9. Github = superpower as a programmer to build new software with unprecedented ease and power, on top of a global base of existing code

  10. AWS = superpower as a programmer to access a global supercomputer with miraculous power on demand for mere dollars

  11. These are only a few examples -- there are probably 100 more new tech superpowers just in the last 10 years of similar magnitude.

  12. I am firmly convinced many people are fundamentally underestimating the power and potential of these new superpowers in the years ahead.

  13. We are only at the very beginning of understanding what people all over the world are going to be able to create and build from here.

  14. As someone who has been in the car service business his whole life, a valuation that high is asking for trouble.

  15. I wish them luck; however, it is a very mom and pop industry and you will be lucky to just make a living.

  16. If Uber becomes the most successful [car service company] out there, you are not looking at a billion dollar company by any means.

  17. I don't know of any competitive advantage Uber has over Carey or Boston Coach...which can launch similar technology in matter of weeks.

  18. I'm looking forward to seeing if Uber is actually successful outside of San Francisco.

  19. Wow, $11 million for a company that is equivalent to calling a taxi! Why didn't I think of that!

  20. I'm surprised. Is Uber in the clear yet legally? Seems like they could still be shut down any day.

A rare interview with one of my heroes: Stanford Professor Donald Knuth, author of Art of Computer Programming...

Written on June 5, 2014, 9:17 p.m.
  1. A rare interview with one of my heroes: Stanford Professor Donald Knuth, author of Art of Computer Programming...

  2. I can speak only to people who happen to have grown up with the strange kind of 'brain organization' that I seem to have...

  3. For lack of a better word, let me simply say that I'm a 'geek.' I haven't got a good definition or a good litmus test for geekhood...

  4. ...but I definitely know it when I see it; and I see it in about 2% of the world's population. The main characteristic is an ability...

  5. ...to understand many levels of abstraction simultaneously, and to shift effortlessly between in-the-large and in-the-small.

  6. Dear young person, if you are a geek, the world needs you, and you will never run out of opportunities to apply your talents.

For the following, substitute rapidly rising tide of products & services for Coke:

Written on June 5, 2014, 3:50 p.m.
  1. For the following, substitute rapidly rising tide of products & services for Coke:

  2. What's great about this country is that America started the tradition where the richest consumers...

  3. ...buy essentially the same things as the poorest. You can be watching TV and see Coca-Cola: you know that the President drinks Coke...

  4. ...Liz Taylor drinks Coke, and just think, you can drink Coke, too. A Coke is a Coke and no amount of money can get you a better Coke...

  5. ...than the one the bum on the corner is drinking. All the Cokes are the same and all the Cokes are good...

  6. .... Liz Taylor knows it, the President knows it, the bum knows it, and you know it. --Andy Warhol, 1975, http://t.co/jvqC0j07Nz

Technology innovation disproportionately helps the poor more than it helps the rich, as the poor spend more of their income on products.

Written on June 4, 2014, 1:28 p.m.
  1. Technology innovation disproportionately helps the poor more than it helps the rich, as the poor spend more of their income on products.

  2. This sounds like it must be a controversial and politically charged position, and yet it is not -- it flows from basic economics.

  3. The best way to understand this is by historical example: What the rich used to have and what the poor now have, due to tech innovation.

  4. Rich have always been able to pay servants to wash dishes; due to tech change, now most US homes have automatic dishwashers.

  5. Rich have always been able to pay servants to wash and dry clothes; now most US homes have automated washers and dryers.

  6. Rich were able to afford to have fresh ice delivered daily to make iceboxes work; now all American homes have refrigerators.

  7. Rich were always able to afford to hire musicians to play in their homes; now audio equipment and digital music are cheap for everyone.

  8. At one point only the rich could pay for horses, buggies, stables, coachmen -- now cars are easily affordable by almost everyone in West.

  9. Go far enough back, only rich could afford hand-copied books or to employ scribes; printing press made books accessible to the poor.

  10. Technology innovation is the main process by which luxury items become produced, packaged, and made affordable for everyone.

  11. Opposing tech innovation is punishing the poor by slowing the process by which they get things previously only affordable to the rich.

  12. And, tech innovation is the process by which everyone in the world will be able to afford things that are plentiful in the West today.

  13. A great lens on this is the US HUD housing survey; shows rapid material progress of poor Americans quite clearly. http://t.co/UkuFNDuUzj

  14. Note that consumer costs rising most quickly (education, health care) have least tech innovation and least market competition.

  15. This is Baumol's Cost Disease: http://t.co/I1QRpLoYor http://t.co/aOBuxn1bVf http://t.co/052AgYy9AB

  16. The way to make health care and education more affordable for more people is more tech innovation, not less. Push onto tech price curve.

  17. if you object not ALL poor can afford product X: The answer is more tech innovation to drive price down further, in every case.

Cycle time compression may be the most underestimated force in determining winners & losers in tech.

Written on June 3, 2014, 7:35 p.m.
  1. Cycle time compression may be the most underestimated force in determining winners & losers in tech.

  2. First clear instance of cycle time compression: Cloud/SAAS vs on-premise enterprise software.

  3. Cloud/SAAS development cycles can be far faster than on-premise software; single instance deployed instantly to all customers.

  4. Further, customers can try and adopt cloud/SAAS far faster than they can try and adopt on-premise software.

  5. Implication: Cloud/SAAS is probably impossible to compete with for on-premise software across multiple product cycles.

  6. Second clear instance of cycle time compression: Product improvement & customer upgrade cycles for phones vs TVs and cars.

  7. Consumers can upgrade their phones every 1-2 years, vs TVs at 5-8 years? Cars at 10-12 years? With phones improving by leaps & bounds.

  8. Implication: At given point in time, your TV can be 4-6 years behind your phone; your car can be 9-10 years behind your phone.

  9. Implication: TVs and cars will become accessories for phones, not the other way around. And already happening: Airplay, Chromecast.

This is probably a good time to say that I don't believe robots will eat all the jobs.

Written on June 3, 2014, 12:51 a.m.
  1. This is probably a good time to say that I don't believe robots will eat all the jobs.

  2. Preceding tweetstream was to extrapolate the idea out all the way, not to make the case that it's what's going to happen.

  3. First, robots and AI are not nearly as powerful and sophisticated as I think people are starting to fear. Really.

  4. With my VC/tech hat on I wish they were, but they're not. Enormous gaps between what we want them to do and what they can do.

  5. So there is still an enormous gap between what many people do in jobs today and what robots and AI can replace, and will be for decades.

  6. Second, even when robots and AI are far more powerful, there will still be many things that people can do that robots and AI can't.

  7. Creativity, innovation, exploration, art, science, entertainment, caring for others... we have no idea how to make machines do these.

  8. Third, when automation is abundant and cheap, human experiences become rare and valuable. Flows from our nature as human beings.

  9. Examples: Price of recorded music goes to zero; live touring business explodes. Price of drip coffee drops; handmade gourmet coffee grows.

  10. You see this effect throughout luxury goods markets. E.g. handmade high-end clothes. Will extend out to far more consumers in future.

  11. Fourth, just as most of us today have jobs that weren't even invented 100 years ago, same will be true 100 years from now.

  12. We have no idea what the fields/industries/businesses/jobs of the future will be; we just know we will create an enormous # of them.

  13. If robots/AI replace people for many of the things we do today, the new fields we create will build on a huge # of people then available.

  14. People 50, 100, 150, 200 years ago would marvel at the jobs that exist today; same will be true 50-100-150-200 years from now.

  15. To argue huge #'s of people will be available but we will find nothing for them (us) to do is to dramatically short human creativity.

The necessary consequence of robots eat all the jobs is everything gets really cheap.

Written on June 2, 2014, 5:49 p.m.
  1. The necessary consequence of robots eat all the jobs is everything gets really cheap.

  2. The main reason to use robots instead of people to make something is when the robot can make it less expensively.

  3. When people can make something that costs less than what robots can make, then it makes economic sense to use people instead of robots.

  4. This is basic economic arbitrage at work. It sounds it must be a controversial claim but it's simply following the economic logic.

  5. So robots eat jobs in field X = products get cheaper in field X = consumer standard of living increase in field X -- necessarily.

  6. So arguing against robots eat jobs is equivalent to arguing punish consumers with unnecessarily higher prices.

  7. Indeed, had robots/machines not eaten many jobs in agriculture and industry already, we would have a far lower standard of living today.

  8. Just as increases in consumer goods prices disproportionately hurt the poor, holding back on robots eating jobs would more hurt the poor.

  9. Same logic applies to trade barriers (import tariffs): disproportionately hurt poor consumers by inflicting higher consumer goods prices.

  10. Here's the arbitrage logic: Suppose humans make widget X profitably at $10 price to consumer. Robots can make X at $5 price to consumer.

  11. Economics drive X to be made entirely by robots; consumers win. But then imagine owner of robots cranks X price to consumer to $20.

  12. Suddenly it's profitable for humans to make X again; entrepreneurs immediately start companies to make X with humans for price $10 again.

  13. Therefore, with rare exceptions, there won't be states where robots eat jobs and products get more expensive. Almost always cheaper.

Thought experiment: Posit a world in which all material needs are provided for free, by robots and material synthesizers.

Written on June 1, 2014, 10:58 p.m.
  1. Thought experiment: Posit a world in which all material needs are provided for free, by robots and material synthesizers.

  2. Housing, energy, health care, food, transportation --> All delivered to everyone for $0, by machines. Zero jobs in those fields remaining.

  3. What would be the key characteristics of that world, and what would it be like to live in it?

  4. First, it's a consumer utopia: Everyone enjoys a standard of living that kings and Popes could have only dreamed.

  5. Fifth, all human time, labor, energy, ambition, and goals reorient to the intangibles: the big questions, the deep needs.

  6. Human nature expresses itself fully, for the first time in history. Without physical need constraints, we will be whoever we want to be.

  7. The main fields of human endeavor will be culture, arts, sciences, creativity, philosophy, experimentation, exploration, adventure.

  8. Rather than nothing to do, we would have everything to do: curiosity, artistic and scientific creativity, new forms of status seeking (!).

  9. Imagine six, or 10, billion people doing nothing but arts and sciences, culture and exploring and learning. What a world that would be.

  10. The problem seems unlikely to be that we'll get there too fast. The problem seems like to be that we'll get there too slow.

  11. Utopian fantasy you say? OK, so then what's your preferred long-term state? What else should we be shooting for, if not this?

  12. Finally, note thought experiment nature of this stream -- this is an extrapolation of ideas, not a prediction for the next 50 years!

  13. Clarification: I'm not talking about Marxism or communism, I'm talking about democratic capitalism to the Nth degree.

  14. I'm not postulating the end of money or competition or status seeking or will to power, rather the full extrapolation of each of those.

The flip side of robots eat all the jobs not being discussed: The current revolution in the means of production going to everyone.

Written on June 1, 2014, 9:47 p.m.
  1. The flip side of robots eat all the jobs not being discussed: The current revolution in the means of production going to everyone.

  2. In the form of the smartphone (and tablet and PC) + mobile broadband + the Internet: Will be in almost everyone's hands by 2020.

  3. Then everyone gets access to unlimited information, communication, education, access to markets, participate in global market economy.

  4. This is not a world we have ever lived in: Historically most people in most places cut off from these things, usually to a high degree.

  5. It is hard to believe that the result will not be a widespread global unleashing of creativity, productivity, and human potential.

  6. It is hard to believe that people will get these capabilities and then come up with... absolutely nothing useful to do with them.

  7. And yet that is the subtext to the this time is different argument that there won't be new ideas, fields, industries, businesses, jobs.

  8. In arguing this with an economist friend, response was But most people are like horses; they have only their manual labor to offer.

  9. I don't believe that, and I don't want to live in a world in which that's the case. I think people everywhere have far more potential.

One of the most interesting topics in modern times is the robots eat all the jobs thesis; best book on topic: http://t.co/2bYZrGgbhr

Written on June 1, 2014, 9:10 p.m.
  1. One of the most interesting topics in modern times is the robots eat all the jobs thesis; best book on topic: http://t.co/2bYZrGgbhr

  2. The thesis is that computers can more and more substitute for human labor, thus displacing jobs and creating unemployment.

  3. At core, this is Luddism (http://t.co/7lZ8vKCXrw) -- lump of labor fallacy, that there is a fixed amount of work to be done.

  4. The counterargument is Milton Friedman: Human wants and needs are infinite; there is always more to do. 200 years of history confirms.

  5. To avoid the Luddite mistake, must believe this time is different, that either (a) there won't be new wants and needs (vs human nature),

  6. Or (b) It won't matter that there are new wants and needs, most people won't be able to adapt to contribute & have jobs in new fields.

  7. While it is certainly true that technological change displaces current work & jobs, and that is a serious issue that must be addressed...

  8. It is equally true, and important, that the other result of each such change is a step function increase in consumer standards of living.

  9. As consumers, we virtually never resist technology change that provides us with better products/services even when it costs jobs...

  10. Nor should we. This is how we build a better world, improve quality of life, better provide for our kids, solve fundamental problems.

  11. Make no mistake, advocating slowing tech change to preserve jobs = advocating punishing consumers, stalling quality of life improvements.

  12. So how then to best help individuals who are buffeted by producer-side technology change and lose jobs they wish they could keep?

  13. First, focus on increasing access to education and skill development -- which itself will increasingly be delivered via technology.

  14. Second, let markets work (voluntary contracts and trade) so that capital and labor can rapidly reallocate to create new fields and jobs.

  15. Third, a vigorous social safety net so that people are not stranded and unable to provide for their families.

  16. The loop closes as rapid technological productivity improvement and resulting economic growth make it easy to pay for safety net.

Economists call it the lump of labor fallacy. It s the idea that there is a fixed amount of work to be done in the world...

Written on May 30, 2014, 5:10 p.m.
  1. Economists call it the lump of labor fallacy. It s the idea that there is a fixed amount of work to be done in the world...

  2. ...so any increase in the amount each worker can produce reduces the number of available jobs.

  3. [Or, equivalently, the adoption of labor-saving technology, or the increased use of offshore labor.]

  4. (A famous example: those dire warnings in the 1950's that automation would lead to mass unemployment.)

  5. As the derisive name suggests, economists view idea with contempt, yet the fallacy makes a comeback whenever the economy is sluggish.

  6. Paul Krugman in 2003, http://t.co/G2YuGi05hb

Consider the development of the tractor, which mechanized virtually all of agriculture over the 20th century...

Written on May 30, 2014, 5:07 p.m.
  1. Consider the development of the tractor, which mechanized virtually all of agriculture over the 20th century...

  2. Somehow new desires and demands sprung up for new kinds of manufactured goods, many of pure entertainment value...

  3. ...and people stayed employed and real wages kept rising. https://t.co/YtvdkmIIiA

Heading into our 5th anniversary Annual Meeting next week, risk and return of high-tech startups and venture capital on my mind...

Written on May 29, 2014, 5:17 p.m.
  1. Heading into our 5th anniversary Annual Meeting next week, risk and return of high-tech startups and venture capital on my mind...

  2. My rough estimate of my personal success (positive vs negative outcome) rate of bets on new products and companies is ~60-65%.

  3. Meaning, of course, my failure (negative vs positive outcome) rate is ~35-40%. I.e. I'm wrong about several things at all times.

  4. Per Tversky/Kahneman loss aversion, the negative outcomes weigh much more heavily on my mind: key challenge = manage own psychology.

  5. Per Taleb antifragility, the saving grace of my business is: Each loss capped at 1x, but wins can scale to 1,000x and even beyond.

  6. I am also blessed in having partners and colleagues over 20 years and today who are often better at picking than I am.

  7. I envy my hedge fund friends who can fully implement strong views weakly held and trade out of bad positions any time they want.

  8. However, because we make hard commitments for 10+ years, we can win from sleeper hits that take longer to develop. That's very satisfying.

  9. On balance, progress is made. But the emotional rollercoaster never stops! Our entire field is constantly exhilarating and terrifying.

  10. The other saving grace is seeing people we work with develop, succeed, and flourish. Enormously fulfilling, makes it all worthwhile.

  11. Commenters correctly point out that for VC (more than other fields), to calculate success rate, must account for great opp'ys passed.

  12. I.e. total at bats = decisions made to pull trigger whether they worked or not, plus decisions to pass on big wins one could have had.

  13. So at bats probably more like 120%-140% of # of decisions made to invest. Knocks overall success rate down a fair amount. Humbling!

  14. Counterargument against counting misses is that success in investing has nothing to do with hitting every good opp'y. But still! :-)

  15. And ultimately, in both startups and VC, success rate (batting average) means nothing; slugging percentage means everything.

Posit: All viral content headlines are simply lies.

Written on May 28, 2014, 4:47 a.m.
  1. Posit: All viral content headlines are simply lies.

  2. You WILL believe what happened next.

  3. Six things about X that you ACTUALLY DON'T need to know.

  4. This will NOT change your life.

  5. Something someone said that will NOT leave you speechless.

  6. A person did this thing and it's NOT weirdly mesmerizing.

  7. The happiest people in the world do NOT actually do this.

  8. However, that playful kitten really is quite cute.

Thesis: Do what you love / Follow your passion is dangerous and destructive career advice.

Written on May 27, 2014, 7:09 p.m.
  1. Thesis: Do what you love / Follow your passion is dangerous and destructive career advice.

  2. We tend to hear it from (a) Highly successful people who (b) Have become successful doing what they love.

  3. The problem is that we do NOT hear from people who have failed to become successful by doing what they love.

  4. Particularly pernicious problem in tournament-style fields with a few big winners & lots of losers: media, athletics, startups.

  5. Better career advice may be Do what contributes -- focus on the beneficial value created for other people vs just one's own ego.

  6. People who contribute the most are often the most satisfied with what they do -- and in fields with high renumeration, make the most $.

  7. Perhaps difficult advice since requires focus on others vs oneself -- perhaps bad fit with endemic narcissism in modern culture?

  8. Requires delayed gratification -- may toil for many years to get the payoff of contributing value to the world, vs short-term happiness.

Since people asked, a quick summary of my own views about economic development over the next several decades:

Written on May 24, 2014, 9:41 p.m.
  1. Since people asked, a quick summary of my own views about economic development over the next several decades:

  2. We need a lot more economic growth for the benefit of everyone, including and especially for lower-income people here and everywhere.

  3. We also need a lot more technological advancement and productivity improvements for the benefit of everyone, again including the poorest.

  4. Further, the good news story of our time is the global rise of billions of people out of poverty. We need this to continue, not reverse.

  5. Theories and policies that push us backwards toward more stasis, stagnation, and statism are both counterproductive and immoral.

  6. None of this is to dismiss the wrenching consequences that individuals may experience during a time of rapid economic change.

  7. For that reason, it is both sensible and moral to couple full-throttle technological capitalism with a strong, rigorous social safety net.

  8. And in fact, rapid economic growth and technological progress is precisely what makes a strong, rigorous social safety net affordable.

  9. We can in fact have our cake and eat it too. Last 500 years have clearly shown that democratic capitalism is a non-zero-sum phenomenon.

A tweetstorm in four parts: The superpowers metaphor for how new technologies enhance human capabilities.

Written on May 23, 2014, 5:21 p.m.
  1. A tweetstorm in four parts: The superpowers metaphor for how new technologies enhance human capabilities.

  2. New technologies can be thought of as giving people superpowers -- superhuman abilities that humans did not have before.

  3. First, let's think about historical technologies through this lens. Starting with man-made fire: superpower to heat and cook things.

  4. Electric lighting = superpower to be active and productive even when the sun is down. AC = superpower to act/produce when too hot.

  5. Steam power and mechanical engines = superpower to exert physical effort in the world way beyond human or animal muscles.

  6. Planes, trains, and automobiles = superpower to travel far faster than feet, animals, or wind could previously carry us.

  7. Telegraph & telephone = superpower to communicate faster and cheaper than messages could be physically carried.

  8. Threshing machines, combines, and tractors = superpower to grow far more food than we could by hand.

  9. And of course there are destructive superpowers: guns and bombs enable killing far more people than we could by hand.

  10. For 500+ years, we've collectively been radically enhancing capabilities of ordinary humans through technology superpowers.

Since people asked, my own strong views weakly held personal stance on Net Neutrality:

Written on May 22, 2014, 5:45 a.m.
  1. Since people asked, my own strong views weakly held personal stance on Net Neutrality:

  2. We need to somehow both retain permissionless Internet innovation & telcos ability to get return on capital for network investment.

  3. My preferred policy route would center on promoting regulation/deregulation and incentives for more last-mile broadband competition.

  4. I see the potential for at least 5-way last-mile broadband competition, at least in non-rural areas:

  5. A: cable, B: telco, C: Google Fiber, D: mobile carrier networks LTE & beyond, and E: Wifi and future derivatives.

  6. There are a whole bunch of things that could accelerate and enhance C, D, and E. We should identify and do those things asap.

  7. One key topic is wireless spectrum; need to get a lot more in the hands of both mobile carriers and into unlicensed classification.

  8. Wifi in particular seems underestimated: If a lot more/different spectrum were available, could go much faster & longer range.

  9. In addition there are a variety of new ideas including satellites/drones, Steve Perlman's DIDO, etc.; we should warmly embrace those.

  10. With sufficient local competition, regulatory pressure much reduced: If one provider plays games, consumers can switch to another.

Topics that anyone who wants to take a position on Net Neutrality and the current FCC issues should thoroughly understand:

Written on May 22, 2014, 5:22 a.m.
  1. Topics that anyone who wants to take a position on Net Neutrality and the current FCC issues should thoroughly understand:

  2. History, technology, and economics of backbones, interconnection agreements, peering, CDNs, caching, colocation.

  3. Current and future telco and cable business models including capex and opex models, rate caps, cost of capital, return on investment.

  4. History, structure, and case law of FCC regulatory framework, including Acts of 1934 and 1996, Title II common carrier rules, et al.

  5. Technology and economics of competitive entrants: Google Fiber, mobile carriers LTE & beyond, Wifi & derivatives, and satellites & drones.

  6. Technology, economics, and culture of permissionless Internet innovation at each layer of the stack, contrasted vs prior platform regimes.

  7. Side note: I don't think I know anyone who understands all of these topics in depth. I know I don't.

When a top exec gets fired & public reasons are unclear/confusing, most common explanation: He/she lost support of his/her direct reports.

Written on May 17, 2014, 6:20 a.m.
  1. When a top exec gets fired & public reasons are unclear/confusing, most common explanation: He/she lost support of his/her direct reports.

  2. It is generally impossible for any board or CEO to leave an executive in place when that exec's direct reports are collectively revolting.

  3. What's interesting: This is true in practice almost regardless of what you think of the reasons for the revolt.

  4. If the reasons for the staff revolt are valid, then clearly a mistake has been made and must be fixed; exec has to go.

  5. If reasons for the revolt are NOT valid, then there's an even bigger cultural problem at that point, with even broader cleanup required.

  6. But even if reasons for revolt not valid, exec under fire is still right in the middle of it, has lost confidence of troops, still goes.

  7. Executives get paid the big bucks because their decisions impact lives and careers of 100s/K's/10K's of employees. Short leash justified.

  8. Plus another theory for high exec compensation is precisely to make it easier to take dramatic action when needed. Implicit safety net.

  9. None of preceding intended to diagnose any specific situation, just the general pattern. Also not excusing any kind of bad behavior.

  10. All of this double-underlines importance of wise governance, leadership development, management training, and cultural integrity.

The great under-appreciated miracle of the iPhone -- couldn't make a reliable phone call for ~3 years yet glorious success anyway!

Written on May 7, 2014, 8:16 p.m.
  1. The great under-appreciated miracle of the iPhone -- couldn't make a reliable phone call for ~3 years yet glorious success anyway!

  2. My explanation: Was so overwhelmingly better at every other function than existing phones, in comparison phone calls didn't matter.

  3. Question: To what extent was that the origin of current phenomenon of users abandoning voice calls in favor of texting + social networks?

  4. The ultimate rope-a-dope marketing strategy. Aha, you thought you were buying a phone? Guess again!!

  5. Could that rope-a-dope marketing strategy work in other fields? Video games? Cars? Banking? Health insurance?

Conventional view of how to value companies: (1) Analyze company + its financials + future cash flows; (2) Calculate correct valuation.

Written on May 7, 2014, 6:02 p.m.
  1. Conventional view of how to value companies: (1) Analyze company + its financials + future cash flows; (2) Calculate correct valuation.

  2. What actually happens: (1) Observe current market valuation; (2) Construct theory and model to explain that valuation.

  3. In this way, George Soros's theory of reflexivity is exactly correct. http://t.co/iX3cWkmZP0

  4. Fundamentals influence prices which influence fundamentals which influence prices which influence fundamentals... ad infinitum.

  5. At cyclical top, high prices drive creation of theories to explain infinite future glory; negative investors & analysts get fired.

  6. At cyclical bottom, low prices drive creation of theories to explain permanent future misery; positive investors & analysts get fired.

  7. Therefore a boom in theories of how everything's a bubble and certain to crash is evidence of a cyclical bottom, not a cyclical top.

  8. Therefore Efficient Market Hypothesis is correct if for all information you substitute all information, theories, noise, and bullsh*t.

  9. Since we are social animals, the challenge of actually standing outside of herd is brutally hard. Pressure to conform is constant/intense.

  10. Famous paper well worth reading: The Limits of Arbitrage http://t.co/rYyZXSaEyd

  11. Another famous paper well worth reading: Noise by the great Fischer Black http://t.co/cjng4qOcQQ

A growing problem I see often now, arguably the ultimate first world problem, but still a problem, and fascinating to watch...

Written on May 7, 2014, 12:50 a.m.
  1. A growing problem I see often now, arguably the ultimate first world problem, but still a problem, and fascinating to watch...

  2. I call it Silicon Valley Syndrome - a high-end applied version of the Paradox of Choice applied to genius high-potential tech whizzes.

  3. Acute strains among Stanford grads and young alumni of biggest hot companies like Google etc. But not limited, can infect anyone here.

  4. Presents as achievement and career paralysis - overload of too many great choices freezes ability to decide, commit, and stick to 1 thing.

  5. Do important work for big co, take 1 of 20 hot startup offers, found own co with any of 5-10 possible cofounders, become junior VC? Agggh!

  6. Not nipped in bud, becomes chronic, damages 5-10 prime career years. Resume becomes saga of job hopping. Once potent potential dissipates.

  7. Years later, friends and colleagues wonder, whatever happened to X? He/she seemed so high potential. Such a shame. Such a waste.

  8. Rarely strikes 18-22. Often starts at 26 after 4 years at hot co. Most common at 26-32. 35+, people either lost forever or come to senses.

  9. Friends of new sufferers encouraged to slap silly, exclaim What's wrong with you? Pick a thing and stick with it! This is not that hard!

  10. Unexpected side effect: More great opportunities open up sooner than they should for new up & comers. Then cycle repeats.

When I was in college (1989-1993) and starting to pay attention to the world, experts and commentators ~all believed two things for sure:

Written on May 4, 2014, 10:08 p.m.
  1. When I was in college (1989-1993) and starting to pay attention to the world, experts and commentators ~all believed two things for sure:

  2. First, Japan was going to utterly own and control the world's technology industry and therefore the world economy.

  3. Everyone knew they had a fundamentally better system: single party government, industrial policy, government-directed economy.

  4. Coupled with Japan's overwhelmingly superior education system, American had zero chance of competing with Japan in any technology area.

  5. Second, America's best days were behind it and my generation (Gen X) would be first to be worse off than our parents.

  6. America was experiencing fundamental and irrevocable cultural, societal, and economic collapse. The die was cast, the results were in.

  7. In retrospect both of these beliefs were artifacts of the (relatively mild) US recession of the late 80s/early 90s + the Japan bubble.

  8. The paranoia peaked in '92 w/Michael Crichton's brilliant novel Rising Sun. http://t.co/XuKhULwFag

  9. By 1996, both of these theories were thoroughly refuted by reality and dropped down the memory hole by most people who held them.

  10. Today, we look back on books like Lester Thurow's Head to Head and Clyde Prestowitz's Trading Places with embarrassment and laughter.

  11. Side note #1: This is not to say that the US doesn't have plenty of challenges; but bottom-of-cycle paranoia is a real phenomenon.

  12. Side note #2: Japan 20yrs later remains amazing nation w/enormous opportunity; I think a glorious economic renaissance is quite possible.

On the other hand, current advantages of mobile native apps vs mobile web, at least as commonly used/deployed today...

Written on May 3, 2014, 6:09 a.m.
  1. On the other hand, current advantages of mobile native apps vs mobile web, at least as commonly used/deployed today...

  2. Apps often have more mobile-native UIs, at least relative to web experiences that have not been extensively adapted for mobile.

  3. Mobile native apps often have better performance than mobile web; meaningful given latency issues with many mobile networks.

  4. Mobile native apps often have better and more complete access to mobile hardware capabilities; meaningful given rapid hardware evolution.

  5. Mobile OSs have acclimated users to having icons for mobile apps on home screen, vs icons for mobile web bookmarks.

  6. Mobile native apps have easy monetization methods that web has historically lacked, including in-app payments & recurring subscriptions.

  7. Mobile native apps have access to OS notification feeds that web experiences don't (or don't easily).

Characteristics of the web that have not been recreated, or have only partly been recreated, with mobile apps so far...

Written on May 3, 2014, 6:01 a.m.
  1. Characteristics of the web that have not been recreated, or have only partly been recreated, with mobile apps so far...

  2. Universal client -- one piece of code on each client device that runs all apps, removing need to install and manage client code.

  3. One-click app access (install) -- run any app by clicking on a link.

  4. Permissionless innovation -- no approval required by anyone to bring new web apps online; no opportunity to bottleneck or censor.

  5. Instant universal updating -- make a server-side change and all running instances of app simultaneously upgraded.

  6. Trivially easy creation of situation-specific apps, w/any programming/scripting language, on any OS, etc. Some just a few lines of code.

  7. Deeply integrated link (URL) model, framework, format, specification -- native to how apps work and how users experience them.

  8. Trivially easy integration and layering of 3rd party services w/web content + apps, from search engines to social networks to Twitter.

Grappling with Piketty: Thesis is 1% will compound capital from here at such a rate as to take all the $ and leave 99% with little?

Written on April 27, 2014, 9 a.m.
  1. Grappling with Piketty: Thesis is 1% will compound capital from here at such a rate as to take all the $ and leave 99% with little?

  2. Underlying assumption seems to be that compounding large pools of capital over long periods of time is straightforward and low-risk.

  3. First question: If that's true, shouldn't national pension programs like US Social Security be immediately privatized & invested same way?

  4. To get twin benefits of harvesting great capital compounding opp'y for benefit of 99%, and also suppressing investment returns of the 1%?

  5. Piketty walks up to this conclusion and then backs away, apparently because that would be too risky for long-term retirement assets?

  6. But if it's that risky for e.g. the Social Security pool, isn't it also riskier than he says for the 1% investing their own capital?

  7. Note that investment approach for SS could be to invest the entire pool like a university endowment, not individuals investing own $.

  8. Second, if capital compounding will work so well for the 1%, isn't that result of world awash with opp'ys to productively invest capital?

  9. And isn't that world the opposite of the secular stagnation/innovation is dead world so many economists believe we are in?

  10. And isn't that world one that is fantastic for consumers who benefit from all of the resulting innovation and technological progress?

  11. And wouldn't Piketty's prescriptions suppress that hypothetical scope and rate of technological and material progress?

  12. So is Piketty therefore proposing that the 99% be made worse off absolutely in order to be better off relatively?

  13. And will the 99% be consulted at any point as to which of these scenarios they'd like to see play out?

  14. Or is the theory that liberal technocratic economists in Paris and places like it get to make those decisions on behalf of the 99%?

Other irony re Piketty: He comes at a time when most pro managers of large long-term institutional $ are deeply worried about the reverse:

Written on April 27, 2014, 12:46 a.m.
  1. Other irony re Piketty: He comes at a time when most pro managers of large long-term institutional $ are deeply worried about the reverse:

  2. A world in which there is a massive surplus of capital relative to opportunities to deploy it productively to compound $.

  3. Approximate thought process of the median professional manager of institutional financial assets today:

  4. Stocks suck, bonds suck, real estate sucks. Hedge funds keep blowing up. Private equity returns reverting to mean. VC on fringe at best.

  5. How can I possibly generate returns sufficient to fund future obligations in a low-growth, low-return world? I am very, very nervous.

  6. Developed world slowing and aging too fast; developing world not developing fast enough. And price of every asset already bid way up.

  7. They and Piketty cannot both be right. Ironically Piketty is the far more optimistic one re future of capitalism, growth, innovation.

I suspect that tech-driven price deflation is the great under-appreciated economic force of our times.

Written on April 24, 2014, 10 a.m.
  1. I suspect that tech-driven price deflation is the great under-appreciated economic force of our times.

  2. If true, both GDP and productivity growth #'s may be decreasingly accurate representations of actual economic progress.

  3. For an example of tech-driven price deflation, consider the physical products that have now been collapsed to $0.99 or free iPhone apps.

  4. Answering machine Barcode scanner Books Calculator Camcorder Camera CD player CDs Compass Credit card terminal

  5. Day planner DVD player DVDs Flashlight Games GPS device Guitar tuner Land line telephone Magazines Metronome

  6. Music player Newspapers Pager Pedometer Radio Remote control Scanner Stereo Stopwatch Television Textbooks

  7. Timer Tuning fork Video game box Video games Voice recorder Walkie talkie Wristwatch

  8. You can undoubtedly add another 100 physical items to this list, probably 1,000, and soon 10,000.

  9. All used to contribute directly to GDP growth and other economic metrics based on $ of production, but not or not as much anymore.

  10. And yet it would be nonsensical to pine for the prior state; the resulting improvement in consumer welfare is obvious.

Prevailing beliefs that I do not share:

Written on April 24, 2014, 9 a.m.
  1. Prevailing beliefs that I do not share:

  2. Tech innovation is dead.

  3. Tech innovation is dead except for the part that will kill all the jobs.

  4. Tech innovation is dead except for the part that will kill all the jobs and give all the money to the 1%.

  5. QE is hopelessly distorting the economy.

  6. Hyperinflation is right around the corner.

  7. Hyperinflation is right around the corner and interest rates are about to skyrocket.

  8. Five billion more people are getting access to the most amazing tools for education, information, creation, and access to markets ever...

  9. ...and yet they will figure out how to do... absolutely nothing with them, and are doomed to lives of spiraling poverty and despair.

  10. To repeat, the preceding tweets in this series are current prevailing beliefs that I do NOT share! :-)

The Smart Person Fallacy in two ease steps. I'm smart, I can learn about situation X and figure out the way it should work.

Written on April 20, 2014, 9 a.m.
  1. The Smart Person Fallacy in two ease steps. I'm smart, I can learn about situation X and figure out the way it should work.

  2. First step: Situation X is likely far more complex with far more moving parts and confused causes and effects than you can imagine.

  3. Autodidacts and polymaths are highly prone to this. True experts in a field often far more skeptical about own ability to understand.

  4. Often suffered by professional writers, journalists, commentators, columnists, analysts, investors ... And venture capitalists.

  5. Second step: In many complex situations, logic matters far less than other factors -- with incentives at the top of the list.

  6. So thinking your way to the answer may well be counterproductive or worse.

  7. Often suffered by intellectuals, academics, theorists, paternalistic left-wingers ... And venture capitalists.

  8. Bonus step: For any mandated change to situation X, unintended consequences are likely to dominate the long term effects.

Third aspect of valuation of tech companies often misunderstood--this time private valuations set by VCs and other private investors:

Written on April 18, 2014, 9 a.m.
  1. Third aspect of valuation of tech companies often misunderstood--this time private valuations set by VCs and other private investors:

  2. This topic was recently explored by @Jessicalessin in this excellent article: https://www.theinformation.com/When-3-Billion-is-Not-3-Billion

  3. A private company in which a sophisticated investor has bought a minority stake for $X/share is not actually worth $X * total # of shares.

  4. First, the entire company has not traded hands, just a small slice of it. So we don't actually know what the whole company is worth.

  5. Second, most financing rounds are for preferred shares, which have special rights. Other shares don't have those rights & are worth less.

  6. Smart VCs think about startup shares less as stock than as options -- options with limited (1x) downside & unlimited (1,000x+) upside.

  7. A share of preferred startup stock ~= A long-dated out-of-the-money call option, paired with a long-dated contingent put option.

  8. The contingent put option is the liquidation preference in preferred stock. Increases odds of getting $ back in a downside sale of the co.

  9. Plus, in some high-valuation late-stage rounds, there are additional downside protections like ratchets, which can be highly valuable.

  10. And, preferred stock brings with it governance rights and information access not available to normal investors. Those have value too.

  11. So you can't extrapolate the value of an entire company from a minority sale of preferred stock. Better just to focus on cash raised.

Second thing often misunderstood about tech valuations: How M&A acquirers decide how much $ to pay for companies they buy.

Written on April 18, 2014, 8 a.m.
  1. Second thing often misunderstood about tech valuations: How M&A acquirers decide how much $ to pay for companies they buy.

  2. The key: Value of a tech company to public or private markets may be completely unrelated to value of same co to a corporate acquirer.

  3. Value of company X to acquirer Y often = Potential impact to acquirer Y's business -- which has a lot more to do with Y than X.

  4. For example, in product businesses, you'll often hear term attach rate -- acquirer Y can attach company X's product to Y's sales engine.

  5. Ex: I sell $20B of servers/year; I buy storage company X doing $100M revenue/year; & I can attach X's product to 20% of my server sales.

  6. Ex cont'd: I can generate new $20B*20% = $4B/year of storage sales attached to my server business. X's standalone revenue is irrelevant.

  7. Ex cont'd: So I can pay up for storage company X based on its projected impact on MY business, way beyond X's independent valuation.

  8. Of course, for the deal to be good, I have to deliver that attach rate. But when it works, and it often does, it's magical & worth doing.

  9. This is literal meaning of attach rate, but there are others -- such as, maybe I know how to better monetize something I buy than they do.

  10. Large $ acquisitions of small co's that seem irrational to outsiders almost always have a rigorous plan like this within the acquirer.

  11. It's just nearly impossible to see from the outside, which is why many outsiders get so confused and upset at the time of acquisition.

  12. But on the other hand, we do not consider it safe for a tech startup to have a plan that DEPENDS on a large acquirer applying this logic.

  13. We only invest in startups that have a plan to be large independent dominant companies on their own, with great businesses in long term.

A few common fallacies about valuation of public and private technology companies:

Written on April 17, 2014, 8 a.m.
  1. A few common fallacies about valuation of public and private technology companies:

  2. First, ask any MBA how to value tech companies, she'll say discounted cash flow, just like any other company: http://en.wikipedia.org/wiki/Discounted_cash_flow

  3. Problem: For new & rapidly growing tech co's, up to 100% of value is in terminal value 10+ years out, so DCF framework collapses.

  4. You can run as many DCF spreadsheets as you want and may get nothing that will help you make good tech investment decisions.

  5. Related to fact that tech co's don't have stable products like soup or brick companies; future cash flows will come from future products.

  6. Instead, smart tech investor thinks about: A future product roadmap/opp'y, B bottoms-up market size & growth, C talent and skill of team.

  7. Essentially you are valuing things that have not yet happened, and the likelihood of the CEO and team being able to make them happen.

  8. Finance people find this appalling, but investors who do this well can make a lot of money, but spreadsheet investing is often disastrous.

  9. Doesn't mean cash flow doesn't matter, in fact opposite: this is the path to find tech companies that will generate tons of future cash.

  10. Corollary: For tech companies, current cash flow is usually useless for forecasting future cash flow--lagging not leading indicator.

  11. This trips up value investors (Prem Watsa!) all the time; tech companies with high cash flows often about to fall off a cliff.

  12. Because current cash flows are based on past products not future products. And profitability often breeds complacence and bureaucracy.

  13. Always, always, always, the substance is what matters: WHO and WHAT. WHO's building the products, and WHAT products are they building.

  14. Brand will not save you, marketing will not save you, channels will not save you, account control will not save you. It's the products.

  15. Which goes right back to the start: Who are the people, what are the products, and how big is the market. That's the formula.

Following up on earlier tweetstream on new tech growth company valuations, here's some current info on large cap tech valuations:

Written on April 15, 2014, 8 a.m.
  1. Following up on earlier tweetstream on new tech growth company valuations, here's some current info on large cap tech valuations:

  2. Since some big tech co's now have gargantuan cash reserves, I like to look at Price/Earnings ratios adjusted for cash on balance sheet.

  3. Apple, the shining jewel of American capitalism, has chinned all the way up to 2014 estimated P/E ex-cash of 11.2.

  4. Google, the company everyone agrees is the General Electric of the 21st century: 2014 P/E ex-cash of 19.4.

  5. Big legacy tech foursome 2014 P/Es ex-cash: Oracle 12.6, IBM 11.1, Microsoft 10.7, Cisco 9.5.

  6. These are still so low as to qualify as generational lows -- public tech P/Es haven't traded this low for this long since the 1970s.

  7. I've said it before and I'll stay it again: If this is a new tech bubble, it's managing to bypass all of the big public tech companies.

  8. So to rationalize all of this, you pretty much have to believe one of three things:

  9. A: This is the weirdest equity bubble ever, ignores the large cap companies that are easy to trade -- not what happened in late 90's.

  10. B: Public market still scarred after 2000 and 2008 crashes, hates tech equities, except a handful of companies delivering rare growth.

  11. C: Or, many large-cap tech companies are in dire trouble, about to be taken apart by new generation of disruptive challengers

Some data on the recent stock market downdraft, assembled by my partner @skupor:

Written on April 14, 2014, 9 a.m.
  1. Some data on the recent stock market downdraft, assembled by my partner @skupor:

  2. For 7 top consumer tech stocks (FB, TWTR, ZU, TSLA, LNKD, P, YELP): median is off -37.4% from highs.

  3. For 4 top enterprise tech stocks (WDAY, SPLK, FEYE, NOW): median is off 42.0% from highs.

  4. NASDAQ overall is off -8.5% from high. Google is off -12.4% from high; Netflix is off -27.8% from high.

  5. 12 tech IPOs in 2014 so far; median is off -22.0% from high but UP 0.3% from offering price (= effectively flat to offering price).

  6. People who want to see a tech bubble look at this data: See! I told you it was a bubble, and now it's crashing.

  7. People who don't want to see a tech bubble look at this data: See! I told you it's not a bubble; market still mostly hates tech.

  8. One cautionary note: This kind of market behavior (big swings up, down, up) often correlates to high short activity.

  9. When shorts take big position, stocks fall. When shorts liquidate, stocks rise. When shorts squeezed out, stocks really rise.

  10. So it can be tricky to try to divine where market wants price to settle in the middle of a vigorous long/short battle, like arguably now.

Saturday's thought experiment: Thinking about the hypothesized corrosive influence of money on politics.

Written on April 5, 2014, 11 a.m.
  1. Saturday's thought experiment: Thinking about the hypothesized corrosive influence of money on politics.

  2. Let's take as given that income inequality is high and rising. Most people have less $ than some people (1 or 10 or something).

  3. Let's also take as given that the mechanism of money corroding politics = Campaign donations --> Advertising, mainly TV commercials.

  4. Presumption is that rich people donate $ to candidates who run TV commercials to persuade poor people to vote against their own interests.

  5. This despite the fact that poor people outnumber rich people, and each person can only vote once. Poor people numerically dominate voting.

  6. We assume TV ads funded by campaign $ from rich people actually convince poor people to vote against their own interests. But do they?

  7. How do we tell difference between poor people getting fooled into voting against their own interests, vs poor people voting how they want?

  8. To believe money has corrosive influence on politics through this mechanism, must we believe that poor people are stupid?

  9. Alternate explanation: Poor people vote how they want, other people don't approve of their choices, and rich people are wasting their $?

  10. How can we tell? Short of assuming that our judgment of how poor people should vote vs how they do vote is prima facie correct?

  11. Meanwhile, how do we feel about people who make top-down judgments about poor people in other political and economic contexts?

  12. If one believes poor people are too stupid to vote in their own bests interests, in what other areas must they also be judged as stupid?

  13. In case it's not obvious: I think poor people (and middle-class people) are smarter than many observers and activists think.

  14. I say that having grown up rural lower-middle-class myself. Poor and middle-class people frequently patronized, know it, don't like it.

  15. I think in many areas of policy we have concluded that poor people need to be protected from themselves; I think this is dangerous.

Twenty years ago today, we started the company called Netscape!

Written on April 5, 2014, 10 a.m.
  1. Twenty years ago today, we started the company called Netscape!

  2. I'd like to mark the occasion by thanking everyone who worked at Netscape and supported Netscape through the years. You all are awesome.

  3. I'm intensely proud of everything you accomplished then and since! I feel like we were able to help make the world a better place.

Something we are seeing more often: Growth-stage private investors in tech startups renegotiating deals *after* signing term sheets.

Written on April 5, 2014, 9 a.m.
  1. Something we are seeing more often: Growth-stage private investors in tech startups renegotiating deals *after* signing term sheets.

  2. Standard SV practice: Investor signs term sheet, does confirmatory Due Diligence, and funds at those terms unless DD surfaces deception.

  3. Standard practice of certain non-SV investors: Signing term sheet is just beginning of terms & price negotiation, which continues 2 close.

  4. This often starts with breathtaking high-valuation term sheet from non-SV investor, convinces co to go exclusive and shut off other talks.

  5. Non-SV investor can then renegotiate deal arbitrarily and with impunity since company may be screwed if that investor walks.

  6. People in SV generally consider this unethical and abusive. Investors from outside SV, though, may consider standard operating procedure.

  7. We are advising our companies to be highly skeptical of outside-SV high-valuation private offers; validate investor ethics carefully.

  8. Further, this practice may hyper-optimize one deal, but that investor gets bad reputation in Valley, impairing ability to do more deals.

  9. Key to win-win: Both company and investor should thoroughly understand each other prior to signing term sheet. Best path to glory.

  10. Corollary: Terms other than valuation matter a great deal. Wide range of abusive/punitive terms exist aside from valuation, do get used.

Two quick Brendan Eich stories:

Written on April 4, 2014, 9:01 a.m.
  1. Two quick Brendan Eich stories:

  2. Brendan was one of the very first Netscape engineers, joined from Microunity, a famous venture-backed chip startup fiasco at that time.

  3. We decided we needed a scripting language for web browsers and web servers. We investigated every option. Brendan said, I can build one.

  4. He single-handedly wrote Livescript, first version in 3 months. Built into browser. Created modern web programming. Borrowed Java syntax.

  5. Sun asked us to embed their new Java into browser; in return we forced them to accept name Javascript for Livescript. They hated that!

  6. At the time I thought Java would dominate client and Javascript would dominate server; turned out reverse happened for 10+ years.

  7. Javascript in browser became BY FAR most widely used programming language in human history. Breathtaking achievement. Beyond amazing.

  8. Years later, after Netscape/AOL and AOL/TimeWarner mergers, Brendan called me to see if I could help free Mozilla into a nonprofit.

  9. I called Jim Barksdale who was on AOL-TW board; with Jim's help, Mitchell Baker and Brendan successfully established Mozilla Foundation.

  10. This was an unnatural act for a big company, could have easily not happened. Mitchell and Brendan made it happen, redefined web again.

  11. So, if you like your browser, your Firefox, and/or your Javascript, whatever your political beliefs, you owe Brendan a debt of gratitude.

  12. Web would be a sparse & barren place without Brendan's work. I can't argue with his decision, but can't wait to see what he does next.