3 ms·

I'm not in favor of the line which the grandparent is arguing but perhaps there's an argument to be made that Uber is, in effect, "eating the seed corn" of the

I'm not in favor of the line which the grandparent is arguing but perhaps there's an argument to be made that Uber is, in effect, "eating the seed corn" of the industry and valuing private gains over public losses. Universities like CMU are responsible for recruiting and training the next generation of great roboticists and there's a difference between responsibly harvesting researchers from an institution in a way that allows the lab to maintain its culture and bounce back vs plundering it in such a way that would take many years to recover from. It's a prisoner's dilemma style situation where all companies involved in robotics would benefit from a thriving academic robotics community, even if it would be more competitive if those researchers were in house. Uber was the first one to publically defect which might cause further defections to the detriment of everyone.

JoeAltmaier 11y ago
Would you stay at CMU while your colleagues are being hired into industry at a multiple of your salary? Yet the argument is, they should stay for some reason. In the end, its up to the researchers to choose their own path. All the companies in robotics making hiring agreements would actually be illegal. Collusion in fixing salaries/blacklisting whole categories of employee candidates is a much worse evil. Maybe we can stop trying to social engineer, and address the situation as it exists.
- shalmanese 11y ago
  But Uber could hire 3 or 4 professors a year, offer internships and fellowships for PhD and Masters students, provide funding grants to professors still working at CMU and slowly work it's way up to a 100 person research lab over the course of 5 years. Or Uber could immediately hire 30 people straight away and destroy the lab and then find itself unable to hire anyone else and stuck at around that number 5 years later. It's better for both CMU and Uber if they perform the former. It just requires Uber to take a longer term view of things rather than optimizing for immediate gain. I'm not arguing that this is the reality on the ground. But I see this as a potential line of argument worth considering.
  - JoeAltmaier 11y ago
    Maybe they'll just move the lab to their own company. They'll have all the researchers after all, with a real budget and real goals.
    - shalmanese 11y ago
      But no institutional capacity to grow new researchers. Producing PhDs is a highly resource intensive endeavor that requires a ton of institutional support. Universities consider it their mission but private companies have no interest in it, even if it's vital to their long term success.

JoeAltmaier 11y ago

Would you stay at CMU while your colleagues are being hired into industry at a multiple of your salary? Yet the argument is, they should stay for some reason. In the end, its up to the researchers to choose their own path. All the companies in robotics making hiring agreements would actually be illegal. Collusion in fixing salaries/blacklisting whole categories of employee candidates is a much worse evil. Maybe we can stop trying to social engineer, and address the situation as it exists.

shalmanese 11y ago
But Uber could hire 3 or 4 professors a year, offer internships and fellowships for PhD and Masters students, provide funding grants to professors still working at CMU and slowly work it's way up to a 100 person research lab over the course of 5 years. Or Uber could immediately hire 30 people straight away and destroy the lab and then find itself unable to hire anyone else and stuck at around that number 5 years later. It's better for both CMU and Uber if they perform the former. It just requires Uber to take a longer term view of things rather than optimizing for immediate gain. I'm not arguing that this is the reality on the ground. But I see this as a potential line of argument worth considering.
- JoeAltmaier 11y ago
  Maybe they'll just move the lab to their own company. They'll have all the researchers after all, with a real budget and real goals.
  - shalmanese 11y ago
    But no institutional capacity to grow new researchers. Producing PhDs is a highly resource intensive endeavor that requires a ton of institutional support. Universities consider it their mission but private companies have no interest in it, even if it's vital to their long term success.