2 ms·

As a PhD student at MIT with many friends at Stanford, I'd like to say that both schools provide an abundance of opportunities for an engineering student to cho

by rywang 16y ago

As a PhD student at MIT with many friends at Stanford, I'd like to say that both schools provide an abundance of opportunities for an engineering student to choose a less specialist path: business school classes, leadership in student organizations, student advocacy, business plan competitions, even industry consulting. As with most universities, the opportunities are there, and your education is what you make of it. As to Rebecca's comments on large-scale projects, when DARPA funding first dried up, the CS and AI Lab (CSAIL) turned towards industrial sponsors and pursued a vision of pervasive computing, incorporating advancements in networking, speech, and NLP. But there has been no shortage of recent DARPA sponsored large-scale projects (e.g. the DARPA Grand Challenge, command and control systems for urban search and rescue, semi-autonomous forklifts for war zone unloading).

hga 16y ago
But are any of those large scale DARPA projects pushing the foundations like some of the prior ones did? The ones you list are very applied, which may be appropriate for a nation at war, I took her point to be "where will the foundation for the next big thing come from?" Hmmm, where would we place their VLSI project (http://en.wikipedia.org/wiki/VLSI_Project http://en.wikipedia.org/wiki/VLSI_Project), which among other things according to Wikipedia supported BSD?
- rywang 16y ago
  IMHO, semi-autonomous robotic control is one of the holy grails of CS. Robotics hardware is maturing rapidly, but the software algorithms for visual sensing, planning and manipulation aren't there yet. This is standing in the way of robots running (vs walking) efficiently, tidying up your cluttered desk, or taking care of our elderly.
  - hga 16y ago
    "semi-autonomous robotic control is one of the holy grails of CS" Agreed 100%, and it's a difficult and worthwhile set of AI problems (just not ones I'm personally interested in :-). I have no trouble with the Grand Challenge given that the state of the art makes it practical, it's that other ... perhaps more foundational stuff is apparently being neglected. The stuff that's "wow!" in 10 years.

by rywang 16y ago

hga 16y ago

But are any of those large scale DARPA projects pushing the foundations like some of the prior ones did? The ones you list are very applied, which may be appropriate for a nation at war, I took her point to be "where will the foundation for the next big thing come from?" Hmmm, where would we place their VLSI project (http://en.wikipedia.org/wiki/VLSI_Project http://en.wikipedia.org/wiki/VLSI_Project), which among other things according to Wikipedia supported BSD?

rywang 16y ago
IMHO, semi-autonomous robotic control is one of the holy grails of CS. Robotics hardware is maturing rapidly, but the software algorithms for visual sensing, planning and manipulation aren't there yet. This is standing in the way of robots running (vs walking) efficiently, tidying up your cluttered desk, or taking care of our elderly.
- hga 16y ago
  "semi-autonomous robotic control is one of the holy grails of CS" Agreed 100%, and it's a difficult and worthwhile set of AI problems (just not ones I'm personally interested in :-). I have no trouble with the Grand Challenge given that the state of the art makes it practical, it's that other ... perhaps more foundational stuff is apparently being neglected. The stuff that's "wow!" in 10 years.