The Business Model Pivot

I saw Zuck’s post on pivoting to private interactions from public posts yesterday and I had a flashback to Bill Gates’s Internet Tidal Wave memo to his company almost twenty-five years ago.

I have always seen a lot of Gates in Zuck. They both have this incredible ability to see someone else’s product and realize that they need to build their own version of it.

But copying someone else’s product is a lot easier than copying someone else’s business model, particularly when you already have a fucking great one that makes you and your shareholders billions of dollars a year.

It will be interesting to watch Zuck do what Gates was ultimately unable to do – completely reboot the company’s business model to position itself to win the next wave in tech.

In the case of Gates, it was the pivot from paid software to free advertising supported software (aka – the attention economy that we are now paying for).

In the case of Zuck, it will be the pivot from monetizing attention to monetizing the protocol. The good thing is he is headed in the right direction, and surrounded some of the smartest people I know in crypto. The bad news is when you have this anchor called a legacy business model, it means making the right moves and making them quickly a lot harder.

Here is an example of one of those choices Facebook will need to make and make correctly:

In any case, it is game on. Being on the verge of 60 years old means I have seen this game play out at least once before and so I have a frame of reference to observe it. That’s really great. It is an exciting time again in tech.

#blockchain#crypto#VC & Technology#Web/Tech

Comments (Archived):

  1. kidmercury

    i don’t see how fb’s coin will be fast enough if it is not on a private blockchain, or a semi-centralized chain like EOS. early adopters will put up with slow apps, but to get the mainstream, speed is required — which tilts the evolutionary bias towards centralization. paradoxically, both centralization and decentralization are simultaneously needed.but the mechanics of how they determine the basket of assets they will peg the coin to, and how they maintain the peg, are the real innovations to watch for in my opinion, as they may suggest how we get closer to the monetary policy revolution.

    1. Twain Twain

      Well let’s see how FB will solve the AI problem. The more centralized the data, the more accurate the predictions.https://uploads.disquscdn.c

    2. Michael Brill

      For now, tokenization > decentralization.

    3. JamesHRH

      It is not paradoxically, its illogically.

      1. kidmercury

        nah, all these fake beefs — men vs women, day vs night, etc — the answer is always both (or neither, for the nihilists). decentralization creates the value, centralization captures it. you can see this in any other protocol, take SMTP for email — open protocol enables email to still be flourishing and still be vital for everyone to be online, but those who captured value and innovated upon it are the centralized email owners. maybe facebook will build a closed token on top of the open ethereum chain, or maybe they will build their own “half open, half closed” type of chain like EOS….either way, they will need both open/closed aka centralized/decentralized, and both must continue to advance simultaneously, in spite of seeming to be opposites.

        1. JamesHRH

          I totally agree with you Kid.But Blockchain is about disintermediation. It does not do both.

          1. kidmercury

            i don’t think blockchain is about disintermediation. mining in proof of work chains is already highly centralized. metamask for ethereum wallets is widespread. disintermediation is only half the story, re-intermediation is the (oft-neglected) other half.

          2. JamesHRH

            If you are taking the middle man out and then putting the middle man back in, then isn’t Blockchain just a new mousetrap?

          3. kidmercury

            sure, though a better one in the sense that it is more scalable. a car, like a horse, is a transportation “intermediary”, but most would agree it is a better one, and that the innovation is meaningful.

          4. JamesHRH

            Hmmmmm.Horse = FedexCar = Internet? = BlockchainI don’t see it.

          5. kidmercury

            more like:horse = debt promises amongst tribescar = monetary policy by central bank/nation-state treasuryairplane = monetary policy by coin + chainsome don’t like the “monetary policy revolution” vision of blockchain, though to me that is by far the most interesting and world-changing part. for those who prefer the more technologically-oriented viewpoint:horse = fax machinecar = internet apps based on tcp/ipairplane = internet apps based on blockchains

          6. JamesHRH

            Horse = fax = copper / dial tone / analogCar = GUI = broadband / tcp-ip / digitalAirplane = ?= broadband / tcp=ip / digital / blockchainI don’t see anything more than an incremental leap.And I don’t kook out on monetary policy being dark web & mainstream at the same time.

  2. Elia Freedman

    I spent the last decade trying to change my own company’s business model and I can tell you, even as a tiny company, it is extremely hard. There is so much inertia around the old model that it took me too long to make the change. It is hard to have the guts to do that when everything is going well. Old customers like what they know and a business model change is all uncertainty.More power to Mark Zuckerberg if he can do it. I left Facebook a few months back but if he truly follows through I’d consider returning. I miss seeing what my family is up to, and watching my cousin’s kids grow up.

  3. William Mougayar

    It is a very defensive move. Facebook is well aware of the threats to its business model. They are very good at sugar coating situations. I think Gates is more sincere than Zuckerberg.

    1. JamesHRH

      Which is why Zuck is more effective. The sociopathy works.

    2. kevando

      It’d be nice if he also copied Apple’s brevity.

      1. William Mougayar

        Yup. Such a long memo.

    3. jason wright

      very good at zucker coating. there’s something not quite right about him. I find it concerning that such a person has so much influence over our lives. i thought the principles of civil society were there to limit such influence. he’s not a democratic representative. he’s a private citizen. it’s dangerous.

      1. William Mougayar

        “zucker coating” good one !

        1. Girish Mehta

          Hi William, seems like Disqus issues continue and my earlier comment is removed as spam…can you take a look ?

          1. William Mougayar

            Un-spammed. Weird, it was caught that way. Sorry about that.

          2. Girish Mehta

            Thanks..maybe because it was too long :-).

          3. William Mougayar

            Not sure. You are white listed noe too.

          4. jason wright

            or just too good.

          5. sigmaalgebra

            Ah, no worries, mate!!! You’ll get used to it! I’ve been spam since grade school!!!! The pretty girls who wouldn’t let me spend money on them for dinner or a movie never called me spam, but it was clear enough that they would have called me much worse!!!

      2. sigmaalgebra

        Ah, the guy knows some German or at least some from The Magic Flute — Zuckerbrot!!

      3. sigmaalgebra

        I make nearly no use of Facebook. Each time I visit, I get a thunderstorm of random full new screens or changes to the current screen. I see no documentation and have little understanding how any of their features or tools work. When I click on things, mostly I get nothing or nothing I wanted.Via the media, I’ve heard this and that bad things about Facebook, politics, privacy, etc., but I never saw anything about politics directly from Facebook and don’t much mind whatever I know about privacy.For politics, one guy there I follow or who is a friend or whatever, posts several times a day and often posts politics, basically a little left of the old John Birch Society and Barry Goldwater, far to the right of Trump. A few times I’ve responded in kind. But I’ve never seen any censorship, etc. from Zuck’s gnomes.Okay, I’ll try to understand what all the objections are: What is supposed to be objectionable about Facebook?Oh, by the way, I wasn’t any good with gossip before Facebook and am not much better now.

        1. jason wright

          The T&Cs are the objectionable. Mining my mind is objectionable. Privatising my thoughts is objectionable.All that we see is all that there is.

  4. Susan Rubinsky

    A vaguely preemptive move on Zuckerberg’s part before the black curtain of regulation changes the game.Imagine the legislators who don’t even know how tech works regulating it? Who will get across the finish line first? Interesting times.Love the Gates reference. If anyone is looking to get a sense of Fred’s frame of reference, I highly recommend watching the series “Halt and Catch Fire.” — fun to watch if you lived through those times –

  5. Tom Labus

    It took a long time for MSFT to get a second life. It also took Bill Gates and Steve Ballmer to be gone too. Satya was the next generation.

    1. JamesHRH

      MS never centralized a control structure. Bill’s attention made you powerful (or your CF stream).So, you cannot steer 8 unruly warlords, even if you created them.Zuck learned from MS and has kept power.

      1. Salt Shaker

        Zuckerberg controls nearly 60 percent of fb stock, and combined w/ insiders controls 70 percent of voting shares. Hard to wrestle him away w/ that much leverage.

        1. JamesHRH

          You could very easily argue that he is not less sociopathic than Bill, it is just that he learned from what Bill did.He is much better strategically.

    2. sigmaalgebra

      Gates was a very bright but also stable enough guy. Ballmer screamed a lot and as far as I could tell had insight no deeper than a fast layer of furniture wax.What does Satya have? I’m not doubting; instead I’d very much like to know but don’t. At this point, Microsoft is a super ship tough to turn, stop, or sink. The C-suite could go on a three year sabbatical without anyone noticing.People are running out of additional really good uses for computers. So far computer applications are for work we already knew very well how to do, and did, manually, and now we need to get beyond that to much more. Cars did that — they were not a “faster horse” and were beyond anything that went before. Airplanes did that.A candidate wide open field for computers is, broadly, providing new, powerful, valuable information we didn’t have before. Computers have been doing this for the technical work in the STEM fields, especially for US national security, but the common man in the street can also use … there is a famous, one word answer to what people want, “MORE”!Here are two really simple, very old things really seriously wrong with Microsoft’s product line:First, Xerox PARC was designing office machines and had an assumption that the work should be via direct, manual manipulation with a GUI (graphical user interface). So, Xerox wanted their office machines to be a lot like a microwave oven — a few buttons to push. Microsoft then assumed that all user computer interactions should be similar.BUMMER. Can’t expect to do good future work in computers using just buttons on a microwave oven. In part it’s like trying to farm 2000 acres using the tools of a flower box, trying to do high end chemistry in the kitchen of an efficiency apartment, trying to build an airplane using just a hammer, saw, file, and sand paper.In particular, the Xerox manual manipulation GUI is super tough to automate: It’s okay for something to be done once or a few times but absurd for 500,000 times. Generally, easily enough to expect, I don’t want manually to manipulate and, instead, want to automate.I want automation. So, sure, since old command lines are MUCH easier to automate than GUIs, I like command lines.My main path to automation is old Rexx. It is not as comprehensive as PowerShell but is much older; I know it very well; I have written hundreds of Rexx scripts; Rexx is darned elegant; the documentation is excellent; and I’m reluctant to go to just PowerShell. Without Rexx, my computing would be like a car without tires. Rexx is one of my very most important, crucial, productive tools.Second, since users type via keyboards, there needs to be software to type into. I type a LOT but for ALL my more important typing I have nothing from Microsoft to type into. Instead, I type into an old, astoundingly powerful, programmable editor KEdit. I have written hundreds of KEdit macros. For me, a computer without KEdit would be a car without wheels.For me and all I do on computers, except just for the Windows operating system and Firefox Web browser, far and away my most important tools are Rexx and KEdit.Microsoft desperately needs to fix this situation. This gap has been obvious for decades. Microsoft is just totally in love with the whole Xerox PARC GUI paradigm — I hated it at the beginning and for anything very important still hate it now. I hate it because it is so wildly awkward and inefficient — it slows me down by a factor of 5-10, and that is a BIGGIE.Over and over and over, Outlook, Excel, Word (I refuse to use and use TeX instead), Microsoft forces me into GUI-based, direct manipulation, one case at a time, with clicking — millions of clicks, maybe millions a week.One of the main things I need to do is to manage my million or so files, growing quickly. So, a standard thing to do is to run a little Rexx program that writes to a file a description of all the files and directories in any subtree of the file system, one file or directory per line. So, per LINE, which means we are already into the land of simple syntax and parsing milk and honey. Next, each line has fields, all fixed length except the last. So, again, the land of milk and honey.Then in KEdit I can select, keep, delete, sort, slice, dice, string search, string change, and at a few a day write little one shot macros to do some task on all 100,000 to 1 million such lines. Then put the result into a file and have a simple Rexx script read the file and do the work — copy, delete, change time-date stamps, search, etc. That’s MUCH better than Xerox PARC and what Microsoft has and is how I get a lot of my work done.E.g., Excedrin headache #498,668,190: Outlook puts my e-mail in a PST file where I have no documentation for the syntax. So, I can’t use my usual KEdit/Rexx approaches. Yes, there’s some old version of Basic that can read and parse a PST file, but last time I looked had to pay for it.For Excel, usually when I want a graph, doing the calculations inside Excel is too darned slow. At times I’ve gotten an Excel file of millions of bytes for a calculation I could do on one page of Rexx, Fortran, C, or Visual Basic .NET. So, for a graph, I calculate outside of Excel and use Excel only for the graphs.Then for graphs in Excel, the defaults seem to have been selected, maybe by some artists, certainly not engineers or experimental physicists, to make the resulting graphs as useless as possible. So, I have to go clicking, a few hundred clicks per graph, just get rid of the defaults and get back to just some simple, useful graphs. I’d like to be able to IMPORT those settings.Next, for 50 graphs, I’d like to be able to AUTOMATE those.Maybe for graphs should give up on Excel and consider Python, R, Mathematrica, etc.If we are going to have KidsWhoCode, then we need to move on and have computers also for PeopleWhoCode and for more, people who do new things in automation and information.KidsWhoCode is good: The idea that end users would code, long would get computer industry middle managers up on their hind legs and angry. But part of the more productive future of computing is more people who can code.

      1. someone

        Ballmer gets no credit for cleaning up the antitrust mess. That took awhile

    3. SFG

      Ballmer was poison. I personally don’t like facebook and so glad to see the Zuck stay at the helm. If they had the right person in there who was not so creepy, who knows how much better facebook could screw with us?

  6. Richard

    Never let a fear mongering crisis go to waste.Facebook takes your dollars – gives you back their greenstamps – and FB makes $ on the float. FB now talks about privacy with a cost while FBs users want nothing but free publicity.The Russians are coming, the Russians are coming.

    1. sigmaalgebra

      Ah, you also saw the connection with greenstamps! Honest, I posted my observation before reading yours!

  7. awaldstein

    I work honestly cause it is the most exciting time available.Such an interesting balance between the depth of perceptions and knowledge that experience brings and the openness that I need to approach each and every new idea and conversation.

    1. Richard

      I think you mean “Honestly, I work”

  8. Vinish Garg

    When I look at the keywords – privacy, decentralization, currency, and trust, the news sounds like a Facebook oxymoron!

    1. Adam Sher

      Right, Gates’ letter addresses something his company was not participating in (internet) and should. A new business venue that Microsoft wants to capture. Privacy etc… is the opposite of FB’s business model, not a compliment or extension. So how does FB square that? It’s almost like if your drug dealer sold heroin and owned the rehab clinics.

      1. ci5er

        That sounds like a great arms-length transaction!

  9. Adam Sher

    I don’t see the parallel b/t Gates and Zuckerberg’s letters. Maybe someone can explain.

    1. sigmaalgebra

      First parallel: They are both rich! Second, they are both in tech. Third, every VC would like to do an A round getting 30% of such a company. Fourth, they both have some really big network effects: Everyone gets on Facebook because everyone else is already there. Everyone uses Windows because all the applications are there, and that is because all the users are there.

  10. Richard

    The government who spent the last 20 years tracking everything is concerned about others tracking anything. And FB steps in as the new Cypto banker to seize on the opprtunity for you to believe you need this. Make politicians appear smart and make the float while doing the same. The Crypto user sheep are such Noobs.

    1. sigmaalgebra

      What’s the big difference between a Facebook use of crypto tokens and the old grocery store “trading stamps”, e.g., Green Stamps?

      1. Richard

        Same pig – different lipstick and pig farmers.

  11. Adam Sher

    FB may be the Karma Hotel (song by Spooks). “It looks like heaven but could be hell…”

  12. Kirsten Lambertsen

    “They both have this incredible ability to see someone else’s product and realize that they need to build their own version of it.”That’s some top shelf shade right there 😀

    1. JamesHRH

      They are both Enneagram 5s, as is Fred.It is not shade, per se.But, yeah..

    2. fredwilson

      Was not meant as shade. Was genuinely meant as a respectful comment

      1. JamesHRH

        Larry Page is pretty good at covering his flanks when he sees someone building a product that threatens the golden goose.

        1. Vasudev Ram

          What are your thoughts about GooglePlus, past and future? When they started it, I read that he gave orders that all products/divisions were to integrate it, or else! (TM) (to the level of saying people who did not, would be let go – IIRC). It did grow some (I mean, big, sure, but not sure how big relative to Google’s other products), maybe because of the Google effect. Now the consumer version is shut down. I did get over a million views on my G+ profile/posts over its lifetime, and the integration of Blogger (where I blog [1]) with G+ was useful, new posts could be automatically shared from my blog (opt-in) to G+. The +1’s were also useful to know people’s reactions to posts. Don’t know how the corporate version is doing / will do. Thoughts?[1]Blog: jugad2.blogspot.comPython posts:…

          1. JamesHRH

            Never used it. Sorry, no comment.

          2. Vasudev Ram

            Got it, NP.

        1. Kirsten Lambertsen

          Some of this is SO Andy Warhol 🙂 (And that’s *not* shade.)

          1. sigmaalgebra

            There you go again, more deep sensitive perceptive social and even artistic insight! I had been assuming that AVC was “nerds only”??But you illustrate one of my favorite points: How is a computer using just key words going to characterize or work with the social, artistic, and literary meaning of content such as your posts? Clearly, bluntly, they won’t.So, what an incongruous juxtaposition: How is a math/computer nerd challenged in things social, artistic, and literary going to write software to do well with the meaning of content good on social, artistic, …, meaning? Hmm, some advanced pure math — nearly only nerds know (more awful alliteration?)

        2. JamesHRH

          He missed the ETH : BTC call, at least by today’s #s!!!

      2. Kirsten Lambertsen

        Yeah, I was attempting humorous irony. I’ve been paying attention long enough to know that you don’t tend to insinuate your criticisms. (It would be excellent shade, though, if intended that way.) Apologies if my humor fell flat!

        1. sigmaalgebra

          Now THAT’S too darned far! And it’s a WOMAN thing — no way can male nerds compete with such depth of social insight and skills with language!

  13. Salt Shaker

    A ploy to mollify rabid dogs. As much PR spin as biz pivot. Did fb really have a choice? Spoken in a language regulators can now understand w/ out the aid of an interpreter.

  14. DJL

    I honestly don’t see how he couldn’t accomplish most of this with some thoughtful features overlaying the FB platform. What they (and the rest) of big-tech needs to fix is censorship. Like several of the FB user comments below his post said “don’t be my Mom and tell me what I should and should not pay attention to.”I did not see this as some watershed business model change. But that is one of the many reasons you are writing AVC and I am scrolling in the rarely-read comments section.

    1. Matt A. Myers

      They’re maximizing for ad revenue, nothing else, and that bias over the years has self-selected and perpetuated to continue that behaviour without consideration for the external costs. This lack of regard for external costs, even as him quoted, was Zuckerberg’s knowing Harvard was building an online version of TheFacebook – along with him saying he doesn’t understand why they were taking so long to launch it, that he could quickly do it: they hadn’t launched yet because the group at Harvard was trying to understanding the privacy, safety, security aspects of it all.

  15. JamesHRH

    To pull another legend out of the mist, you should remember that Jack Welch put Scott McNealy from SUN on his BoD. He then sucked Scooter’s brain dry re: web + tech and paid him back in prestige.Jack said that startup people are nuts, because they only have one chance to be right and get it right, whereas GE could be wrong and get it wrong multiple times without going under. He then tried to prove that point by making Immelt his successor.There is no way banks are going to roll over the way media did. They are online already and 100x fiercer.Plus, a FB stable coin isn’t a dis intermediated financial system, it is just FB replacing banks and sovereign fiat currency with FB & crypto protocols.

  16. jason wright

    Why the ‘Zuck’ diminutive? It’s ingratiating and sycophantic in tone. Facebook is a threat to open society. The name is Zuckerberg, and he needs to be formally evaluated.”They both have this incredible ability to see someone else’s product and realize that they need to build their own version of it.” Far too generous. They steal ideas from others. Incredible indeed.

    1. fredwilson

      Fair question. Wasn’t conscious. But I did write it that way

      1. jason wright

        The Zuckerborg syndrome? Subconscious assimilation. Fight it.Big is not beautiful. Small is Beautiful – E.F. Schumacher.…”As if people mattered”.

        1. JamesHRH

          Familiarity breeds shorthand.What’s Beyonce’s last name?Hell, what is Be! short for?

          1. jason wright

            ‘Z’ – Imagine that.

        2. Mac

          Would that be called a Zucker Punch?

    2. sigmaalgebra

      Yup, I’ve long wondered how products that can be so easily copied can be the basis of a really good business for either the company with the original version or the copy. E.g., right away, for Google, Bing had a good copy and IIRC there are also good copies in each of Russia and China.I know; I know; can also be successful with brand, network effects, etc.Gee, guys, I thought that one of the more important keys to business success was something really good but tough to copy due to some Buffett moat from patents, trade secrets, a large list of customers that get major upgrades each year or so, etc. Of these sources of a moat, to me, the key was a technical core difficult to understand, duplicate, or equal. Heck, that’s DARN it, much of exactly WHY I stopped my good career, going well, annual salary 6 times what a new, high end Camaro cost, and got a Ph.D. in what looked like a Niagara Falls source of such secret sauce (awful alliteration). Or, my career was in applied math and computing for US national security, and there was secret sauce on the table for three meals a day and plenty of snacks, and, feeling an intense need to learn more, I was dripping in secret sauce and eager for more — hence the Ph.D.At the time, much of the “secret” was that not many people understood the sauce yet: E.g., at the time the fast Fourier transform (FFT) was a hot topic, along with what to do with it in digital filtering in time and space, etc. and a guy wired an FFT box. He proudly showed it to me. I asked him about how his circuitry handled the different addressing at each of the stages, and he confirmed that that was the hard part. Then I ruined his day: I told him about a version of the FFT that had the same addressing at each of the stages that, then, would have saved him nearly all the work, complexity, and chip count in his box. I suspect that eventually he picked up his jaw from the floor.That FFT stuff was published, and I’d only read it. There can also be good stuff not published and harder to guess than that special version of the FFT.There are some big gaps in the STEM fields and associated markets: Nearly all the brain power and burned glucose is in academic research publishing papers nearly no one reads, mostly even if the papers are really good, looking for tenure and especially research grants, right, as in money. Awkward, indirect way to make money! But, maybe appropriately enough, the grants are for relatively far out stuff; in important respects, that’s where, long term, for our country, the big gains are.But here is the gap: Can also do research, say, based on some of the best of the far out academic research, that has more direct commercial value AND take the steps to obtain that commercial value, and due to the gap, that research is seriously neglected. So, for a moat, for intellectual property difficult to duplicate or equal, get rows of barbed wire from (i) not many people know the far out academic research, (ii) the people who do know, nearly always quite bright and hard working, are culturally and in their interests a long way from money making applications, (iii) there can be more research to do, and (iv) the workers, managers, BoD members, etc. in the commercial world don’t have even a weak little hollow hint of a tiny clue how to duplicate or equal, are like native tribes in the Amazon jungle seeing water vapor in the sky from passenger jets flying over at 30,000 feet up!E.g., this stuff that Google did a good version and now Redmond, China, and Russia can right away make good copies does not have to hold.So, with good secret sauce difficult to duplicate or equal, to make a copy, can’t just throw developers at it; even a really good electronics engineer didn’t see the version of the FFT he needed; I got our software house “sole source” on some work because I knew how to do a good job with power spectral estimation and exploitation (thank you J. Tukey), and he and his colleagues didn’t — not even research and from just reading the literature. Can’t throw it at the computer science faculties of Stanford, Berkeley, CMU, or MIT because of the prerequisites — not in computer science and a bit much for the computer science profs and grad students to study. Similarly, can’t just throw it at CEOs or BoDs because they can’t hire the needed people.The people with the money just believe that with a few extra dollars they can get anything done in STEM field technology. Well, often that’s true, but it doesn’t have to be and can be wildly false! Even if the false cases are rare, well, for the intended gains that’s goodness!Well, for PCs, Intel borrowed heavily from work done at Multics and IBM and made it cheap! For an operating system, Microsoft also borrowed heavily from the same two. For a user interface, Microsoft borrowed heavily from Xerox PARC and X-Windows. TCP/IP? That came from ARPA Net and some quite good, earlier work, and maybe the first implementation was on Unix at Berkeley, as BSD — Berkeley Software Distribution.That way, first, Microsoft got rid of the typewriters. Later they nearly got rid of office paper. Now they have gotten rid of a lot of humans at service desks and bank teller windows and a lot of USPS paper mail! Google? An electronic version of an old library card catalog key word subject index. Facebook? Gossip! Amazon? Well done, on-line, electronic catalog of easy to ship products — e.g., lots of DVDs but no ice cream!Uh, information is important for business, entertainment, and more. Now the main location of information is content on the Internet. The good content on the Internet is enormous, and the total content is some factors of 10 larger. What people want is the content with the meaning they want. Okay, how to give them that — uh, safe for work, families, children, please? A problem: How to solve the problem of the meaning of the content? The computer science departments may say that that is a Holy Grail problem that presumes good work in handwriting recognition, voice recognition, natural language syntax, semantics, and understanding, actual machine intelligence, and, finally, for the Holy Grail, meaning. Well, you see, there’s some quite far out pure math where …! Where software developers, computer science, Sand Hill Road, Redmond, etc. have no chance!

      1. jason wright

        “The good content on the Internet is enormous, and the total content is some factors of 10 larger.” It’s still a discovery issue. I still think it’s about finding good people and following their footprints.

        1. sigmaalgebra

          Sure, an engine for search, discovery, recommendation, curation, subscription, etc. has to discover the URLs somewhere! Then the hard work starts on meaning! Occasionally for some purposes, sure, footprints can work — “Hey, Joe, did you see this PDF on ESOPs?”, but no matter how good Joe’s friend Sam is, there’s a lot more to a good solution for Joe no matter how much Sam knows!

    3. Matt A. Myers

      Likewise, they have arguably the most valuable dataset in the history of humankind – if they actually had a creative visionary in the lead then they could have been consistently and exponentially improving society.

  17. Girish Mehta

    I was in the PC industry when the MS pivot you refer to happened. There is a big, big difference between Gates’ pivot and Zuckerberg’s announcement. Gates’ move in 1995 recognized a market reality and moved quickly. It was Not driven by the regulator/government considerations. In fact, the reverse – the subsequent Microsoft anti-trust case was a reflection of how successful Microsoft was in an aspect of that pivot which was integration of IE with the OS.Zuckerburg’s move is damage control and is response to the scrutiny the company is facing both from the government and the market. They are doing well financially, but at this point Facebook has little room left to avoid the elephant in the room that is privacy.Consider the context to Microsoft’s move. Microsoft announced the pivot to the world on Pearl Harbor Day 1995. In the issue of Infoworld dated just 3 days before – Dec 4 1995, Bob Metcalfe predicted that the Internet would collapse in 1996.Now, Metcalfe was not nobody – he invented Ethernet, gave the world “Metcalfe’s law”, founded 3Com which in 1995 was the pre-eminent networking company (Cisco’s glory years were still a couple of years way). I remember talking to many analysts in the first 2 weeks of Dec 1995 on the contrast – One of the most respected voices in networking then was predicting that the Internet would collapse the next year, and one of the most powerful companies in the industry was announcing an all-out focus on the internet. Both less than a week apart.Microsoft succeeded with the pivot. This wasn’t about pivoting from paid software to free advertising supported software. You are missing the counter-factual…the past is also a probability distribution (whose one outcome came true).Microsoft’s pivot was about stopping Sun-Netscape at that point. Sun was leading the ABM (Anything But Microsoft) effort, they had a brand that said “The Network is the Computer”, and the had this shiny, new thing called Java. McNealy was brilliant in telling stories and pithy one-liners – the Sun story was of Javastations running JavaOS with a browser and Java applets, connecting to Sun servers at the back end. McNealy said why do you need a full client OS and applications – that’s like everybody owning a independent power plant in their own home instead of using metered electricity. That was a strike at the heart of Microsoft’s business. But Sun really got the pieces of their solution together only by the end of 1996 and Microsoft had already moved before then. The rest is history. The Sun network computer vision did not play out. Sun had a few good years until 2000 with their servers, then struggled to recover after the dot-com bust, and finally got acquired.

    1. Amar

      <disclosure> I have no professional success to offer that gives me any credibility to question Mark Zuckerber’s choices or his business success. But that should not stop one from calling out extremely condescending business practices. I am a product manager who thinks all product managers should truly grapple with the ethics of their choices </disclosure>From 2015:… On Tuesday, Mark Zuckerberg, the company’s founder and CEO, told an internal corporate Town Hall that Facebook would soon be rolling out the feature.“I think people have asked about the dislike button for many years,” Zuckerberg said, according to Re/Code and Business Insider. “Today is a special day because today is the day I can say we’re working on it and shipping it.”Zuckerberg added that the “Dislike” button wouldn’t work like Reddit’s up- and downvote mechanism, which is probably the most famous “dislike” button on the Internet. Instead, he said, users would now have a technical way to express regret or sympathy in a situation.“What they really want is the ability to express empathy. Not every moment is a good moment,” he said.Really, !!!! Mark -> you discovered the notion that “not every moment is a good moment?” … The arrogance here is such a condescending FU to the rest of the educated world.Facebook did not invent emotion, people knew how to relate, love, care and support each other before facebook and will continue to do so long after facebook is done. Yes, Facebook absolutely made the world smaller. But … at some point it became 100% clear that all Facebook cared about is seeing the rest of the world’s attention as fossil fuel for their stock. There is ruthlessness and focus which I commend but I don’t see them as innovators creating new business models.Bill Gross and (I guess someone before them) made ppc real. Google and Facebook succeeded in optimizing inventory (discovering new forms of inventory) to keep the ppc beast fed and happy. I am sure FB the stock is a tremendous success.From my vantage point: MSFT’s innovation literally powered the world and created a tremendous amount of positive externalities. In my opinion FB is at best neutral and can be argued has created more negative global externalities than positive.disclosure: i do not own FB stock 🙂

      1. Matt A. Myers

        From my observations over the years relating to Facebook, from its founding and so on, I’m not sure Mark has access to his empathy – his reactions and understanding more seem like learned logic; there’s a simple hearing test to check for imbalances that could actual confirm if he has development blocks relating to this.

        1. Amar


          1. Kesha

            Affiliate marketing world job opportunities increasingly becoming a emerging trend all over the over world nowadays. A current research suggests more than 72% of the people are working in internet based job opportunities at their house without the difficulties. Everyone wishes to spend some time daily with his/her best friends by going any specific wonderful place in the country or any other country. So web-based earning enables you to complete the work at any time you want and enjoy your life. Though picking the best strategy and also establishing the right destination is our ambition toward achieving success. Already the majority of people are generating such a fantastic profit of $34000 in one week by using suggested as well as outstanding ways of make money online. You will start to get paid from the first day when you look at our web site. >>>>>

    2. cavepainting

      The Microsoft bet on browser was not a pivot of a business model by any means, but a misplaced bet on a future outcome (that browser somehow would become the gateway to the next wave of IT investments) that did not come to pass. The opportunity turned out to be in search, social, and later in mobile – all waves that Microsoft missed by some distance.

      1. sigmaalgebra

        My guess was that with Internet Explorer (IE), Microsoft was on the way with a good idea: With their software on both the client and server side, they would “own both ends of the wire”: All the clients would use IE because it worked well with all the Microsoft server side software, and all the server people would use Microsoft’s server side software because it worked well with all the clients.Then about each 3 months, update the software at both ends of the wire to obsolete everything else.Sounded powerful to me.But as part of this owning both ends of the wire, they came out with ActiveX, and it was one gigantic, could drive a fleet of 18 wheelers through, security hole.What did work was just something simple, Firefox, with just HTML and CSS and before JavaScript. And now for the big difference, we have JavaScript.Microsoft made the same mistake IBM did: Both counted on the end user having a LOT of local expert technical support. I know; I know; the need for so much technical support wasn’t a flaw, it was a feature, a dependency creating, revenue generating, please the third party consultants feature.Then some people noticed that mostly what individuals were doing with Windows was usually quite simple, especially e-mail and Web browsing. Maybe with PCs easier to use, people would have done more, but as things were, they didn’t. Soooo, go for laptops, pads, smartphones, and KEEP IT SIMPLE. In the case of Apple, make it like jewelry. Also add audio and video both playback AND recording. GPS. Take advantage of the geometry of a small camera to have a small f/number but still great depth of field. I still want only to use a desktop, but Apple, Android, Samsung, and more made a big BUNDLE.

      2. Girish Mehta

        That is specifically why I said “in an aspect of the pivot… since that browser integration was the focus of the anti-trust case. But Microsoft’s focus was more than the browser as you can see in Gates’ memo.The correct description of what Microsoft did (and this was common terminology) would be “Embrace and Extend” rather than pivot.It was never a bet on a future outcome around the browser per se, it was about co-opting the internet. It is not Microsoft vs the Internet, it is Microsoft and the Internet. And thereby having an answer to Sun-Netscape.Postscript – Microsoft missed many things. This is about what they did not miss. Because the Sun story about a different form of computing was not completely wrong, in at least its abstraction level . A different version of that story emerged many years later under a new name – Cloud computing. Different in the specifics, but similar in its abstraction – remember McNealy’s comment about why would you build an electric power plant in each home. And Microsoft today has a $25 Billion cloud business. That came about in the Satya era but you have to be alive to take advantage of the new opportunity when that emerges. Sun was a technology juggernaut, it is nowhere today.

    3. William Mougayar

      “Damage control” – exactly that. Well said.

  18. iggyfanlo

    If FB wants to get it right, they should add @reedhastings to their BoD. To me, truly the most AMAZING example of business model pivot ever. Even to the point of completely siloing the two business groups and telling the legacy one that they were on a long term ice float.

  19. Jim Peterson

    Isn’t pivot normally used to describe models not working? Facebook is absolutely killing it and Instagram is the next wave. It’s the default choice of advertisers (plus Google). Pricing is rising… Zucks just looking for future waves-I have no love for Facebook (except the stock). The service reminds me of Don Henley’s “Working It.”“But, the barons in the balcony are laughingAnd pointing to the pitThey say, Aw look, they’ve grown accustomed to the smellNow, people love that shitAnd we’re workin’ it “But the business is amazing.

  20. cavepainting

    As Facebook as a public social network enters an era of slow/declining growth, this is the natural pivot to make. But there is serious deficit of trust in the Facebook brand. And Zuck is hard to take seriously after all that has come to pass.Whatsapp is the asset of the future. It can possibly become the Wechat of India and other emerging markets outside PRC with focus on payments, commerce, and becoming the primary gateway to the internet. The $19B acquisition looks like a steal at the moment.

  21. sigmaalgebra

    Ah, pivots, what about pivots for the better in international First Ladies!!Men only department: Did you see the video of the Oval Office visit today from the Czech Republic and, there, the two First Ladies????? Double WOW!First Lady Melania Trump and Czech Republic first lady Monika Babišová!!!!Uh, Disqus has some new but really serious bugs in their JavaScript they use to handle the standard multi-line text boxes.

  22. MickSavant

    excellent use of fuck

  23. Terry J Leach

    Facebook business model change will ultimately lead to the end of it’s monopoly on social. Reading Benedit Evan’s tweets and blog post titled, “Microsoft, Facebook, Trust and Privacy”, I see parallels between Microsoft and the rise of SaaS and between Facebook and widespread adoption of Blockchain applications due to network effects. Facebook billions of users will be introduced to coins and current ecosystem of blockchains and exchanges will incorporate Facebook’s coin and extent with their own coins which are beyond Facebook’s control. Coinbase will be huge winner. Any organization and group can exchange value based on a stable coin. Facebook by issuing cyptocurrency will set in motion the demise to it’s walled garden.

  24. JamesHRH

    Fred is invested in someone who has bet their company on that idea – Kik. Its not happening.