4 ms·

I always wondered, if you're an executive with perfect foresight and your analysis shows that you could either cling to your legacy technology and make a billio

I always wondered, if you're an executive with perfect foresight and your analysis shows that you could either cling to your legacy technology and make a billion dollars for 10 years and then zero dollars after that as the inevitable disruption happened or go all in on self-disruption and make 100 million dollars until the end of time, which would really be the better choice? Given that disrupting yourself would have a 100 year payoff period, it seems far better to try and squash any attempts at change, both internally and externally until progress becomes inevitable. Look at the music industry for example, even a few extra years of CD dominance would exceed the profits of all digital music sales ever. Of course, if you go this route, your incentive is to play clueless and out of touch since even giving credence to the disruptive idea hastens it's arrival. I'm sure there were a lot of executives who were genuinely clueless but this at least shows how otherwise intelligent executives would have a strong incentive to act clueless.

lyonlim 10y ago
Well, some forward looking companies trying to avoid being disrupted now really invest in or set up completely separate subsidiaries to explore this new areas of growth. They even go so far as to have completely different management and buildings. The upside to this is there's no existing bias in management decisions for the new company. Instead, it offers them a potential insider look, and understanding of this new technology..and when the time comes for a pivot, they have something solid to build on.
jwatte 10y ago
Intel is next. They're making billions on big, complex, expansive CPUs. Someone like ARM /will/ replace them, at a much smaller dollar volume.
- frankchn 10y ago
  I am not so sure, mainly because a lot of Intel's advantage comes from its process technology. The ARM instruction set might well take over, but you still need someone to make them. Even at the current 14/16nm nodes, Intel's process is significantly smaller than TSMC or Samsung's [1]. I suspect we will see consolidation (a la GloFo/Samsung) and Intel branching out into being a foundry, but with building 10nm++ fabs running into the billions of dollars, I don't see Intel going anywhere any time soon. [1]: http://newstechnow.com/plans-launch-14nm-finfet-technology-chip-in-2015 http://newstechnow.com/plans-launch-14nm-finfet-technology-c...
  - majewsky 10y ago
    > The ARM instruction set might well take over looks around himself It hasn't already?
- collinmanderson 10y ago
  The Windows world (and therefore most corporate offices) are very much dependent on the x86-64 architecture. The ecosystem will need to switch to web or (ARM) apps for Intel to die. My eye doctor's office is switching to Android tablets, so maybe it's possible.

lyonlim 10y ago

Well, some forward looking companies trying to avoid being disrupted now really invest in or set up completely separate subsidiaries to explore this new areas of growth. They even go so far as to have completely different management and buildings. The upside to this is there's no existing bias in management decisions for the new company. Instead, it offers them a potential insider look, and understanding of this new technology..and when the time comes for a pivot, they have something solid to build on.

jwatte 10y ago

Intel is next. They're making billions on big, complex, expansive CPUs. Someone like ARM /will/ replace them, at a much smaller dollar volume.

frankchn 10y ago
I am not so sure, mainly because a lot of Intel's advantage comes from its process technology. The ARM instruction set might well take over, but you still need someone to make them. Even at the current 14/16nm nodes, Intel's process is significantly smaller than TSMC or Samsung's [1]. I suspect we will see consolidation (a la GloFo/Samsung) and Intel branching out into being a foundry, but with building 10nm++ fabs running into the billions of dollars, I don't see Intel going anywhere any time soon. [1]: http://newstechnow.com/plans-launch-14nm-finfet-technology-chip-in-2015 http://newstechnow.com/plans-launch-14nm-finfet-technology-c...
- majewsky 10y ago
  > The ARM instruction set might well take over looks around himself It hasn't already?
collinmanderson 10y ago
The Windows world (and therefore most corporate offices) are very much dependent on the x86-64 architecture. The ecosystem will need to switch to web or (ARM) apps for Intel to die. My eye doctor's office is switching to Android tablets, so maybe it's possible.