3 ms·

I just watched a series of videos by Technology Connections of the opposite playing out at RCA leading to the failure of the CED video system. They hired a bunc

I just watched a series of videos by Technology Connections of the opposite playing out at RCA leading to the failure of the CED video system. They hired a bunch of smart people with various science/engineering degrees from prestigious universities, gave them essentially free reign to research things, and all they ended up doing was floundering around for decades, not wanting to develop anything to practicality because that's less interesting than new theoretical research and is less likely to get them more published papers. Seems like you certainly need at some degree of direction and limitations on research teams.

MattGaiser 6y ago
The problem there is more that they chose researchers to do engineering tasks. There are plenty of people inclined to build, just not those people.
dialamac 6y ago
Have seen those videos (quite good) and very familiar with the downfall of RCA, but not sure I agree with this take. RCA basically set up an unworkable scenario and management did not heed the sunk cost fallacy to realize they were down the garden path. A stylus based cheaply made consumer electronic device with the technology at the time was pretty much impossible (and then of course by the time tech had advanced a stylus based system was totally moot). You could have hired manhattan project people and it wouldn’t have changed the physical/economic realities of what they were trying to do.
- jimmaswell 6y ago
  I got the impression everyone on the research team was more interested in publishing papers and going down obscure research than more immediately useful results. That kind of research has its value but I think you need more direction and a mix of more people that are more interested in making working stuff.
smabie 6y ago
The ultimate example of this is Symbolics, the company behind the General Lisp machine. PhDs controlled the place and they ended up wasting gargantuan sums of money on building the "perfect" Lisp machine. It's still probably the most advanced OS ever designed and yet it failed miserably. Like all things, balance is required. Engineers and finance people need to be on equal footing, to serve as a system of checks and balances. The when either group becomes to strong, it spells doom. It's usually the finance types that win these struggles of power though (I'm not exactly sure why), so don't have as many examples of engineers destroying companies. Though, I think Google might be a good example of flaky engineers taking over and the bad that results from it.
- p_l 6y ago
  The big financial fuckups at Symbolics were done by non-PhD "business people" that got into company later on, especially certain big inflexible contracts on real estate. This turned out to be a giant anchor when a proper war chest would have allowed finishing Ivory replacement and/or actually delivering full scope of OpenGenera (version 2.0 doesn't even cover what was planned for 1.0)

MattGaiser 6y ago

The problem there is more that they chose researchers to do engineering tasks. There are plenty of people inclined to build, just not those people.

dialamac 6y ago

Have seen those videos (quite good) and very familiar with the downfall of RCA, but not sure I agree with this take. RCA basically set up an unworkable scenario and management did not heed the sunk cost fallacy to realize they were down the garden path. A stylus based cheaply made consumer electronic device with the technology at the time was pretty much impossible (and then of course by the time tech had advanced a stylus based system was totally moot). You could have hired manhattan project people and it wouldn’t have changed the physical/economic realities of what they were trying to do.

jimmaswell 6y ago
I got the impression everyone on the research team was more interested in publishing papers and going down obscure research than more immediately useful results. That kind of research has its value but I think you need more direction and a mix of more people that are more interested in making working stuff.

smabie 6y ago

The ultimate example of this is Symbolics, the company behind the General Lisp machine. PhDs controlled the place and they ended up wasting gargantuan sums of money on building the "perfect" Lisp machine. It's still probably the most advanced OS ever designed and yet it failed miserably. Like all things, balance is required. Engineers and finance people need to be on equal footing, to serve as a system of checks and balances. The when either group becomes to strong, it spells doom. It's usually the finance types that win these struggles of power though (I'm not exactly sure why), so don't have as many examples of engineers destroying companies. Though, I think Google might be a good example of flaky engineers taking over and the bad that results from it.

p_l 6y ago
The big financial fuckups at Symbolics were done by non-PhD "business people" that got into company later on, especially certain big inflexible contracts on real estate. This turned out to be a giant anchor when a proper war chest would have allowed finishing Ivory replacement and/or actually delivering full scope of OpenGenera (version 2.0 doesn't even cover what was planned for 1.0)