12 ms·

End of Cycle?

163 points by akharris 10y ago

jsprogrammer 10y ago
It seems like whenever I hear of cycles, no example past "the last" (~2000) is given. What evidence is there that these are regularly recurring cycles?
- tostitos1979 10y ago
  If you are a student of history, you start seeing these things. I was in tech in 1995, and at the start of the Internet boom it was not obvious to everyone what was happening. You should read Bill Gate's "The road ahead". He (and many others) predicted things like Netflix and the information super highway but they got the delivery mechanism dead wrong - people were convinced it was going to be over cable TV networks. I've heard similar things about the boom/bust days of Commodore in the mid-80s. I agree with the general thesis that (a) people are searching for the next big thing, and (b) we'll likely get it wrong a few times.
  - basseq 10y ago
    The challenge is when people start to try and predict the next cycle and what it will look like. Yes, you can get the broad brush-strokes right, but details are usually very wrong and timing is way off.
    - reysc 10y ago
      Glad that painting is here to stay.
  - selmat 10y ago
    Have you any good resources for recommendation?
  - nickt 10y ago
    In Kevin Kellys new book, The Inevitable [1], he talks about the trends and long-term forces that push and pull the actual implementations and delivery mechanisms of new technology. I'd suggest that understanding the themes narrow down the search area for the Next Big Thing. [1] https://www.goodreads.com/book/show/27209431-the-inevitable https://www.goodreads.com/book/show/27209431-the-inevitable
  - amyjess 10y ago
    > He (and many others) predicted things like Netflix and the information super highway but they got the delivery mechanism dead wrong - people were convinced it was going to be over cable TV networks. Oh yes, I remember how around 1994 people were predicting the rise of "interactive TV" that would take over the world in 1997. The descriptions of what "interactive TV" would do were very similar to the web.
    - VLM 10y ago
      The immediate predecessor of that was the multimedia CDROM. Its all the same idea and same skills of course. The course curricula for a BSCS with a specialization in multimedia in 1993 is identical to webdev in 2016 except for obvious language syntax differences. Graphic arts/design classes from the art school, marketing classes from the biz school, a couple programming classes, etc.
      - tostitos1979 10y ago
        Ha .. almost forgot that one. Do you remember Encarta? Wikipedia of the old days :)
  - dboreham 10y ago
    Um...for the most part it IS delivered over cable tv networks.
- droopyEyelids 10y ago
  It's all based on the idea of The Business Cycle https://en.wikipedia.org/wiki/Business_cycle https://en.wikipedia.org/wiki/Business_cycle People apply that idea to how tech works. You're right to question the evidence for it. The business cycle itself is just a bunch of ideas and theories people use to try and make sense of a chaotic, unpredictable system that is never the same twice. Sometimes it's useful, sometimes it's not. And "theory" in this case refers to the layman's idea of theory. Not the scientific meaning like the "theory of gravitation".
- nostrademons 10y ago
  You could argue that 2008-2009 was actually the beginning of the last cycle. Aside from including an economic crash (which usually, but not always, accompanies a shift in the tech cycle), it also included a shift in delivery mechanisms from web to mobile, and a shift in predominant business types from social, content-driven businesses (Web 2.0) to marketplaces and on-demand services. Before 2001-2003 was 1991-1993, which (aside from the recession) featured intense investor interest in VR, tablet/pen computing, WebTV, etc. That period of uncertainty ended with Netscape's introduction and the rise of the WWW. There was a mini-shift around 1985-86, with the introduction of Windows and Steve Jobs's ouster from Apple. That changed the paradigm of desktop apps from 8-bit PC apps hand-coded in assembly to more slick GUI apps written in C or Pascal that used OS frameworks. Like the other cycles, it was also accompanied by a lot of hype and malinvestment, but largely in the B2B world: the big buzzwords in 85/86 were AI, 4GLs, and 3D computer graphics. Finally, the PC revolution started in the mid 70s. This also had a similar period of uncertainty, coupled with a recession - take a look at old magazines from the late 60s and early 70s. But instead of computing, the hot technology areas then were flying cars, supersonic transports, and home appliances. Makes me wonder if the next big thing might be something not computer related at all, perhaps aviation (again), rocketry, or material science.
  - zjaffee 10y ago
    Also going deeper into this, one of the common trends is huge liquidity events which allow there to be a huge emergence of venture capital funds. The growth of tech stocks in the late 90's gave people tons of money to blow, and since they had trust in the tech economy since had made them tons of money, they further invested into tech. Notice how massive growth of companies such as Google, Apple, Microsoft, and Amazon over the past 5 years. Additionally notice how many of the big VC's of today were early investors in companies such as FB. This gave them the capital and trust they needed to invest in all these random startups. I bet that an Uber IPO will give VCs the money they need in order to continue to invest in more and more companies. The success of silicon valley is that every 8-10 years, there is a new company that reaches unprecedented valuations, and in turn people reinvest that money into the next generation of companies who will reach those same levels.
- amyjess 10y ago
  Back in college, a professor of mine talked about cycles and used the NASA boom of the '60s as an example. When it ended, engineers just abandoned their houses in Cape Canaveral and moved back to their hometowns. I wish I could remember more of that lecture; all I remember is him describing a 7-year boom/7-year bust cycle with the NASA example as an example of how suddenly a boom can end.
- ganeumann 10y ago
  Not proof that this happens every time, but some evidence from the 1980s: http://reactionwheel.net/2015/01/80s-vc.html http://reactionwheel.net/2015/01/80s-vc.html
- axplusb 10y ago
  I would recommend the book Technological Revolutions and Financial Capital: The Dynamics of Bubbles and Golden Ages [0] by scholar Carlotta Perez. She studies longer cycles though (~50 years) but the mechanism she describes seems pretty sound for the matter at hand. It goes more or less like this: - capital in search for long term returns goes to early moves of a big technological shift - as successes from the new technology get more and more apparent, it attracts a much larger slice of capital available, and eventually gets over-funded (the real opportunity of this technology is limited) - a bubble forms, most capital is in for a quick speculative return - back to square one with a new technology (and former bubble bursts) [0] https://www.amazon.com/Technological-Revolutions-Financial-Capital-Dynamics/dp/1843763311 https://www.amazon.com/Technological-Revolutions-Financial-C...
  - dredmorbius 10y ago
    An excellent if obscure book. Interesting to find another fan. I'm not convinced she's right, I think she's more right than most, and she's definitely interesting and well-researched.
- xapata 10y ago
  It's very easy to confuse an autoregressive time series with a wave.
- dredmorbius 10y ago
  Cyclical behavior in business and economics has been noted for a long time, certainly since the Industrial Revolution, and arguably before. I think it's very dangerous to speak of cycles of predefined lengths, which recur regularly. I think it is useful to look at dynamics, how they occur, how they differ, the specific technological mechanisms driving these, and what the future implications are. The questions are deep, vast, perplexing, and far from settled. Carlotta Perez's book is mentioned here, I second that. Paul Mason's Postcapitalism argues based on Kondratiev cycles, and has a lot of strong anecdotal data though I question Mason's economic instincts. Robert Gordon's The Rise and Fall of American Growth is an excellent look backward at the period 1870-1970, and onward to the present, though I think his view forward is somewhat clouded. It's not that I disagree with his headwinds and projections of harder times ahead, it's that he's missed some far more fundamental drivers and dynamics. Alvin Toffler's Future Shock also deals in technology cycles and their dynamics. I've only skimmed bits but suspect that would be insightful. There's a bunch of crud written on the topic as well, but if you're interested in the question, you're not going to be short of content.
Animats 10y ago
The section under "People move from bits to atoms without realizing the change in underlying fundamentals" is important. He writes (condensed): "Many tech investors are shifting from investing in bits-driven business (software) into atoms driven businesses (anything you need to manufacture). ... investors are applying tech multiples expectations to these radically different types of businesses. This is unlikely to end well. ... "If you are merely using software but the business fundamentals have not shifted - than the startup ... will not merit tech multiples. A software-enabled, network connected, crowd funded, smart toaster is, when all is said and done, still just a toaster."* Birchbox. Soylent. Maybe Uber. You can think of others. (Uber may be a bubble. They've raised and spent a huge amount of money underselling competitors. It's rumored that only in SF is Uber profitable.) (Amazon succeeded because they figured out how to do online retail really well. Remember "one click ordering"? (The key to that is not the one-click ordering. It's that the user can easily cancel for a short period after ordering. That makes one-click ordering safe for the user. Most retailers still don't get this.) Then they developed new distribution technology. Amazon has 30,000 Kiva robots and their order pick time is down to 15 minutes at some distribution centers. They're now in the same business Webvan had 15 years ago, but their costs are low enough that it works.)
- colechristensen 10y ago
  >They're now in the same business Webvan had 15 years ago, but their costs are low enough that it works. People frequently mistake Amazon's business. They're not a store or a delivery service, Amazon is a logistics company. What they do at the frontend is important, but their value as a business is everything in between... warehouses, picking systems, delivery logistics, AWS... doing as much of that stack themselves as possible and doing it very efficiently is their core. Just try to buy AAA batteries to convince yourself that being a store isn't Amazon's core competence. (the review system, the recommendation engine, the product organization, it's all mediocre at best)
  - VLM 10y ago
    The future of massively computation intensive personal level logistics is the amazon drone delivers your daily packet of stuff which contains four AAA batteries because amazon has figured out your remote control battery which is currently working, should be dead by this evening. Also it'll have two rolls of TP and one of your favorite breath mints and one squeeze packet of ketchup because it knows you're having french fries with dinner tonight. Personal logistics is going to look a lot more like the star trek replicator (although enormously higher latency... for now) than like a web page store. The blind faith that "... on the internet" will make you a millionaire, will be replaced by a blind faith in the infallibility of personal logistics. Why if Amazon is sending my wife baby clothes, no need to even bother taking a pregnancy test because how could Amazon ever be wrong, etc.
Animats 10y ago
On the cycle front, we've hit the end of ad-supported startups. Google and Facebook now get most ad revenue. I mentioned previously that only two of YC's current crop are ad-supported. Incidentally, we passed "peak Twitter" some time ago. Look at TWTR stock. We may have passed "peak social". We've hit the end of "apps". There's an app for everything, and most of them are losing money now. We ought to be hitting the end of "online retail is 'tech' and deserving of high multiples". So what's next? There's a lot of stuff in the pipeline that's promising, but hard to do. AI. Robots. Automatic driving. Automatic driving is close, unless someone screws up big-time. AI is going to chew up a lot of desk jobs over the next decade. Robots are still hard, but we're now hearing about larger robot deployments than ever before. Amazon has 30,000 robots. Foxconn is installing 60,000. We've never seen single customer numbers like that before in robotics. Vision processing will be big. Lots of semi-intelligent systems will be looking at video and making decisions. From store inventory to crime detection, there's a role for intelligent video processing. VR? Probably not. VR looks to be the next 3D TV. It's just too much of a hassle to go mainstream.
- nemaar 10y ago
  > VR? Probably not. VR looks to be the next 3D TV. It's just too much of a hassle to go mainstream. I disagree, I think VR is going to be the next big thing but it will take a few generations of HW and SW to get there. VR mixed with real time video of your surrounding (augmented reality) is a new environment for brand new applications but it needs to reach the state where it is hidden in your average looking glasses to be really mainstream.
  - josh2600 10y ago
    "I disagree, I think VR is going to be the next big thing but it will take a few generations of HW and SW to get there". I would say, based solely on this statement, that VR is not the next big thing according to your own narrative (as the next big thing will likely be within one to two hardware generations).
  - wpietri 10y ago
    I share his skepticism of VR as a major business. At this point we have a 150-year history of people being wowed by 3D and mistaking novelty for utility. That includes the Brewster stereoscope in the 1850s; the early 1940s wave that gave us the ViewMaster; the anaglyph 3D movies from the 50s to the 80s; the first VR wave of the 90s; and the 3D movie and TV boom/bust of the last decade. These all had significant excitement and commercial success. The Brewster stereoscope was introduced to great fanfare at the Victorian Exhibition of 1850. It eventually sold 250,000 units. The ViewMaster has sold millions of viewers and 1.5 billion reels. (In WWII, the US military even bought 100k viewers for training.) All of these products are now marginal or extinct. Novelty aside, none of them is much better than the 2D alternative. I'm not saying VR will never work. But I am saying that anybody inclined to make a big financial bet on VR should think very hard about this history. Generations of smart entrepreneurs and their customers have mistaken 3D's cool factor for lasting utility. In the end, everybody has gotten bored, shrugged, and gone back to 2D. This is true even when, as with 3D movies, the hassle is as small as putting on a pair of glasses, something millions of people otherwise do daily.
palakchokshi 10y ago
With vision processing, augmented reality and the internet of things I think we are primed to live in a world where devices will be personified in AR. Personified devices will then be able to tell me what they are capable of and I will be able to query them to learn about them. I will also be able to find out what's wrong with them. I think chatbots will be the ideal interface to talk to IoT devices for most things however AR will be needed for a richer experience/in depth device instruction. Envision a world where a google glass like device take see what you are seeing, identify what it you are looking at, figure out if it is an IoT device, notify you in AR if it is (via icon hovering over device or something), allow you to "activate" the device's personified self and talk to it, question it, assign a task to it or a variety of other interactions supported by the IoT device.
- rimantas 10y ago
  And what problems will it solve?
  - AnimalMuppet 10y ago
    Well, a number of years ago, my wife asked me when a building had been built. I thought, "Just click on Help/About and look at the copyright date... Oh, wait. I can't do that on a physical object." But with VR, I could. Now, I'm not saying that I think that VR is the next big thing. I would pay approximately $0 for the ability to know when a building was built, and I've only ever wanted to know (even mildly) once.
    - ec109685 10y ago
      Zillow and done.
      - AnimalMuppet 10y ago
        On a commercial building?
  - Nydhal 10y ago
um_ya 10y ago
Cryptocurrencies are the next cycle.
emilong 10y ago
> The period of 2004 to the 20-teens will be viewed as the era of... the lean startup. Does this mean that the success of lean approaches will not extend to next cycle? I.e. that they're only applicable to the current "bits-based" businesses?
- Swizec 10y ago
  I know a few people who run IoT startups. When creating a new version of your piece of hardware involves shipping the designs off to China, getting the hardware back, checking that it's correct (repeat until actually correct), and then pulling "Yes, mass produce" ... before you even have anything consumer facing. Basically, each "iterate" takes a few months and many thousands of dollars. Lean kind of stops being all that lean. Lean is easy in software. Thinking before doing is important in hardware.

163 points by akharris 10y ago

It seems like whenever I hear of cycles, no example past "the last" (~2000) is given. What evidence is there that these are regularly recurring cycles?

tostitos1979 10y ago
If you are a student of history, you start seeing these things. I was in tech in 1995, and at the start of the Internet boom it was not obvious to everyone what was happening. You should read Bill Gate's "The road ahead". He (and many others) predicted things like Netflix and the information super highway but they got the delivery mechanism dead wrong - people were convinced it was going to be over cable TV networks. I've heard similar things about the boom/bust days of Commodore in the mid-80s. I agree with the general thesis that (a) people are searching for the next big thing, and (b) we'll likely get it wrong a few times.
- basseq 10y ago
  The challenge is when people start to try and predict the next cycle and what it will look like. Yes, you can get the broad brush-strokes right, but details are usually very wrong and timing is way off.
  - reysc 10y ago
    Glad that painting is here to stay.
- selmat 10y ago
  Have you any good resources for recommendation?
- nickt 10y ago
  In Kevin Kellys new book, The Inevitable [1], he talks about the trends and long-term forces that push and pull the actual implementations and delivery mechanisms of new technology. I'd suggest that understanding the themes narrow down the search area for the Next Big Thing. [1] https://www.goodreads.com/book/show/27209431-the-inevitable https://www.goodreads.com/book/show/27209431-the-inevitable
- amyjess 10y ago
  > He (and many others) predicted things like Netflix and the information super highway but they got the delivery mechanism dead wrong - people were convinced it was going to be over cable TV networks. Oh yes, I remember how around 1994 people were predicting the rise of "interactive TV" that would take over the world in 1997. The descriptions of what "interactive TV" would do were very similar to the web.
  - VLM 10y ago
    The immediate predecessor of that was the multimedia CDROM. Its all the same idea and same skills of course. The course curricula for a BSCS with a specialization in multimedia in 1993 is identical to webdev in 2016 except for obvious language syntax differences. Graphic arts/design classes from the art school, marketing classes from the biz school, a couple programming classes, etc.
    - tostitos1979 10y ago
      Ha .. almost forgot that one. Do you remember Encarta? Wikipedia of the old days :)
- dboreham 10y ago
  Um...for the most part it IS delivered over cable tv networks.
droopyEyelids 10y ago
It's all based on the idea of The Business Cycle https://en.wikipedia.org/wiki/Business_cycle https://en.wikipedia.org/wiki/Business_cycle People apply that idea to how tech works. You're right to question the evidence for it. The business cycle itself is just a bunch of ideas and theories people use to try and make sense of a chaotic, unpredictable system that is never the same twice. Sometimes it's useful, sometimes it's not. And "theory" in this case refers to the layman's idea of theory. Not the scientific meaning like the "theory of gravitation".
nostrademons 10y ago
You could argue that 2008-2009 was actually the beginning of the last cycle. Aside from including an economic crash (which usually, but not always, accompanies a shift in the tech cycle), it also included a shift in delivery mechanisms from web to mobile, and a shift in predominant business types from social, content-driven businesses (Web 2.0) to marketplaces and on-demand services. Before 2001-2003 was 1991-1993, which (aside from the recession) featured intense investor interest in VR, tablet/pen computing, WebTV, etc. That period of uncertainty ended with Netscape's introduction and the rise of the WWW. There was a mini-shift around 1985-86, with the introduction of Windows and Steve Jobs's ouster from Apple. That changed the paradigm of desktop apps from 8-bit PC apps hand-coded in assembly to more slick GUI apps written in C or Pascal that used OS frameworks. Like the other cycles, it was also accompanied by a lot of hype and malinvestment, but largely in the B2B world: the big buzzwords in 85/86 were AI, 4GLs, and 3D computer graphics. Finally, the PC revolution started in the mid 70s. This also had a similar period of uncertainty, coupled with a recession - take a look at old magazines from the late 60s and early 70s. But instead of computing, the hot technology areas then were flying cars, supersonic transports, and home appliances. Makes me wonder if the next big thing might be something not computer related at all, perhaps aviation (again), rocketry, or material science.
- zjaffee 10y ago
  Also going deeper into this, one of the common trends is huge liquidity events which allow there to be a huge emergence of venture capital funds. The growth of tech stocks in the late 90's gave people tons of money to blow, and since they had trust in the tech economy since had made them tons of money, they further invested into tech. Notice how massive growth of companies such as Google, Apple, Microsoft, and Amazon over the past 5 years. Additionally notice how many of the big VC's of today were early investors in companies such as FB. This gave them the capital and trust they needed to invest in all these random startups. I bet that an Uber IPO will give VCs the money they need in order to continue to invest in more and more companies. The success of silicon valley is that every 8-10 years, there is a new company that reaches unprecedented valuations, and in turn people reinvest that money into the next generation of companies who will reach those same levels.
amyjess 10y ago
Back in college, a professor of mine talked about cycles and used the NASA boom of the '60s as an example. When it ended, engineers just abandoned their houses in Cape Canaveral and moved back to their hometowns. I wish I could remember more of that lecture; all I remember is him describing a 7-year boom/7-year bust cycle with the NASA example as an example of how suddenly a boom can end.
ganeumann 10y ago
Not proof that this happens every time, but some evidence from the 1980s: http://reactionwheel.net/2015/01/80s-vc.html http://reactionwheel.net/2015/01/80s-vc.html
axplusb 10y ago
I would recommend the book Technological Revolutions and Financial Capital: The Dynamics of Bubbles and Golden Ages [0] by scholar Carlotta Perez. She studies longer cycles though (~50 years) but the mechanism she describes seems pretty sound for the matter at hand. It goes more or less like this: - capital in search for long term returns goes to early moves of a big technological shift - as successes from the new technology get more and more apparent, it attracts a much larger slice of capital available, and eventually gets over-funded (the real opportunity of this technology is limited) - a bubble forms, most capital is in for a quick speculative return - back to square one with a new technology (and former bubble bursts) [0] https://www.amazon.com/Technological-Revolutions-Financial-Capital-Dynamics/dp/1843763311 https://www.amazon.com/Technological-Revolutions-Financial-C...
- dredmorbius 10y ago
  An excellent if obscure book. Interesting to find another fan. I'm not convinced she's right, I think she's more right than most, and she's definitely interesting and well-researched.
xapata 10y ago
It's very easy to confuse an autoregressive time series with a wave.
dredmorbius 10y ago
Cyclical behavior in business and economics has been noted for a long time, certainly since the Industrial Revolution, and arguably before. I think it's very dangerous to speak of cycles of predefined lengths, which recur regularly. I think it is useful to look at dynamics, how they occur, how they differ, the specific technological mechanisms driving these, and what the future implications are. The questions are deep, vast, perplexing, and far from settled. Carlotta Perez's book is mentioned here, I second that. Paul Mason's Postcapitalism argues based on Kondratiev cycles, and has a lot of strong anecdotal data though I question Mason's economic instincts. Robert Gordon's The Rise and Fall of American Growth is an excellent look backward at the period 1870-1970, and onward to the present, though I think his view forward is somewhat clouded. It's not that I disagree with his headwinds and projections of harder times ahead, it's that he's missed some far more fundamental drivers and dynamics. Alvin Toffler's Future Shock also deals in technology cycles and their dynamics. I've only skimmed bits but suspect that would be insightful. There's a bunch of crud written on the topic as well, but if you're interested in the question, you're not going to be short of content.

Animats 10y ago

The section under "People move from bits to atoms without realizing the change in underlying fundamentals" is important. He writes (condensed): "Many tech investors are shifting from investing in bits-driven business (software) into atoms driven businesses (anything you need to manufacture). ... investors are applying tech multiples expectations to these radically different types of businesses. This is unlikely to end well. ... "If you are merely using software but the business fundamentals have not shifted - than the startup ... will not merit tech multiples. A software-enabled, network connected, crowd funded, smart toaster is, when all is said and done, still just a toaster."* Birchbox. Soylent. Maybe Uber. You can think of others. (Uber may be a bubble. They've raised and spent a huge amount of money underselling competitors. It's rumored that only in SF is Uber profitable.) (Amazon succeeded because they figured out how to do online retail really well. Remember "one click ordering"? (The key to that is not the one-click ordering. It's that the user can easily cancel for a short period after ordering. That makes one-click ordering safe for the user. Most retailers still don't get this.) Then they developed new distribution technology. Amazon has 30,000 Kiva robots and their order pick time is down to 15 minutes at some distribution centers. They're now in the same business Webvan had 15 years ago, but their costs are low enough that it works.)

colechristensen 10y ago
>They're now in the same business Webvan had 15 years ago, but their costs are low enough that it works. People frequently mistake Amazon's business. They're not a store or a delivery service, Amazon is a logistics company. What they do at the frontend is important, but their value as a business is everything in between... warehouses, picking systems, delivery logistics, AWS... doing as much of that stack themselves as possible and doing it very efficiently is their core. Just try to buy AAA batteries to convince yourself that being a store isn't Amazon's core competence. (the review system, the recommendation engine, the product organization, it's all mediocre at best)
- VLM 10y ago
  The future of massively computation intensive personal level logistics is the amazon drone delivers your daily packet of stuff which contains four AAA batteries because amazon has figured out your remote control battery which is currently working, should be dead by this evening. Also it'll have two rolls of TP and one of your favorite breath mints and one squeeze packet of ketchup because it knows you're having french fries with dinner tonight. Personal logistics is going to look a lot more like the star trek replicator (although enormously higher latency... for now) than like a web page store. The blind faith that "... on the internet" will make you a millionaire, will be replaced by a blind faith in the infallibility of personal logistics. Why if Amazon is sending my wife baby clothes, no need to even bother taking a pregnancy test because how could Amazon ever be wrong, etc.

Animats 10y ago

On the cycle front, we've hit the end of ad-supported startups. Google and Facebook now get most ad revenue. I mentioned previously that only two of YC's current crop are ad-supported. Incidentally, we passed "peak Twitter" some time ago. Look at TWTR stock. We may have passed "peak social". We've hit the end of "apps". There's an app for everything, and most of them are losing money now. We ought to be hitting the end of "online retail is 'tech' and deserving of high multiples". So what's next? There's a lot of stuff in the pipeline that's promising, but hard to do. AI. Robots. Automatic driving. Automatic driving is close, unless someone screws up big-time. AI is going to chew up a lot of desk jobs over the next decade. Robots are still hard, but we're now hearing about larger robot deployments than ever before. Amazon has 30,000 robots. Foxconn is installing 60,000. We've never seen single customer numbers like that before in robotics. Vision processing will be big. Lots of semi-intelligent systems will be looking at video and making decisions. From store inventory to crime detection, there's a role for intelligent video processing. VR? Probably not. VR looks to be the next 3D TV. It's just too much of a hassle to go mainstream.

nemaar 10y ago
> VR? Probably not. VR looks to be the next 3D TV. It's just too much of a hassle to go mainstream. I disagree, I think VR is going to be the next big thing but it will take a few generations of HW and SW to get there. VR mixed with real time video of your surrounding (augmented reality) is a new environment for brand new applications but it needs to reach the state where it is hidden in your average looking glasses to be really mainstream.
- josh2600 10y ago
  "I disagree, I think VR is going to be the next big thing but it will take a few generations of HW and SW to get there". I would say, based solely on this statement, that VR is not the next big thing according to your own narrative (as the next big thing will likely be within one to two hardware generations).
- wpietri 10y ago
  I share his skepticism of VR as a major business. At this point we have a 150-year history of people being wowed by 3D and mistaking novelty for utility. That includes the Brewster stereoscope in the 1850s; the early 1940s wave that gave us the ViewMaster; the anaglyph 3D movies from the 50s to the 80s; the first VR wave of the 90s; and the 3D movie and TV boom/bust of the last decade. These all had significant excitement and commercial success. The Brewster stereoscope was introduced to great fanfare at the Victorian Exhibition of 1850. It eventually sold 250,000 units. The ViewMaster has sold millions of viewers and 1.5 billion reels. (In WWII, the US military even bought 100k viewers for training.) All of these products are now marginal or extinct. Novelty aside, none of them is much better than the 2D alternative. I'm not saying VR will never work. But I am saying that anybody inclined to make a big financial bet on VR should think very hard about this history. Generations of smart entrepreneurs and their customers have mistaken 3D's cool factor for lasting utility. In the end, everybody has gotten bored, shrugged, and gone back to 2D. This is true even when, as with 3D movies, the hassle is as small as putting on a pair of glasses, something millions of people otherwise do daily.

palakchokshi 10y ago

With vision processing, augmented reality and the internet of things I think we are primed to live in a world where devices will be personified in AR. Personified devices will then be able to tell me what they are capable of and I will be able to query them to learn about them. I will also be able to find out what's wrong with them. I think chatbots will be the ideal interface to talk to IoT devices for most things however AR will be needed for a richer experience/in depth device instruction. Envision a world where a google glass like device take see what you are seeing, identify what it you are looking at, figure out if it is an IoT device, notify you in AR if it is (via icon hovering over device or something), allow you to "activate" the device's personified self and talk to it, question it, assign a task to it or a variety of other interactions supported by the IoT device.

rimantas 10y ago
And what problems will it solve?
- AnimalMuppet 10y ago
  Well, a number of years ago, my wife asked me when a building had been built. I thought, "Just click on Help/About and look at the copyright date... Oh, wait. I can't do that on a physical object." But with VR, I could. Now, I'm not saying that I think that VR is the next big thing. I would pay approximately $0 for the ability to know when a building was built, and I've only ever wanted to know (even mildly) once.
  - ec109685 10y ago
    Zillow and done.
    - AnimalMuppet 10y ago
      On a commercial building?
- Nydhal 10y ago

um_ya 10y ago

Cryptocurrencies are the next cycle.

emilong 10y ago

> The period of 2004 to the 20-teens will be viewed as the era of... the lean startup. Does this mean that the success of lean approaches will not extend to next cycle? I.e. that they're only applicable to the current "bits-based" businesses?

Swizec 10y ago
I know a few people who run IoT startups. When creating a new version of your piece of hardware involves shipping the designs off to China, getting the hardware back, checking that it's correct (repeat until actually correct), and then pulling "Yes, mass produce" ... before you even have anything consumer facing. Basically, each "iterate" takes a few months and many thousands of dollars. Lean kind of stops being all that lean. Lean is easy in software. Thinking before doing is important in hardware.