* Your education is important to the discussion. *
It's only important to you, and only because it's the only measure you have by which to assign the value of others. Hell, just 128 days ago, you were begging for advice on which chocolates to get your fellow employees so they would like you. You don't have experience, but you sure talk a good game.
* Care to comment on this? * Get into a hot startup and listen and learn. The first six months will provide you with more experience than you'd get in a PhD program at any university (short of MIT or Stanford). *
Absolutely.
The first six months of a hot startup is a microcosm or activity. Designs and ideas flying, deals, market tests, focus groups, venture capital pitches, startup camps, all night code binges, and more. It's 20 hours on, 4 hours off, for months at a time, pure adrenaline and pipelined learning at its best.
While you could argue that the VOLUME of information in a PhD program is greater, it takes six to eight years, and the VAST majority of that information is forgotten and never used. A bright person could easily pick up experience in a startup that they wouldn't see for ten years post-grad, if they just get engaged and pay attention.
Short of Stanford, MIT, and to some extent RIT, there really aren't many actively entrepreneurial schools that promote technology development from within. That's why I said "you can always go back". Sure, most choose not to, but that's a personal problem.
He should take a year off to work at one or more startups, then go back to school, if school is what he wanted to do.
So, let's take a look at Cornell, one of the top CS schools (#5 I think).
Here's their "Practicum in Operating Systems" (CS 4411 Project 1) for Fall 2010:
http://www.cs.cornell.edu/courses/cs4410/2010fa/CS4411/proje...
Now, the same course in 2009:
http://www.cs.cornell.edu/courses/cs4410/2009fa/CS4411/proje...
See any difference? Any modifications? Any enhancement or change of any sort?
Nope. None.
It was the same in 2008, but the page has been removed.
Two years of time has passed, new OS architecture designs have been published, some implemented, new threading paradigms documented, and some are in production, but the exact same project, with the same code, same requirements, and same final problem is to be solved, by the same morons that have no incentive to innovate because they're tenured.
Sure, you could argue "it's the basics, dummy!"... but wait...
It's a SENIOR YEAR project, and they STILL have these poor drones working on the food service project. This is AT BEST a sophomore project.
Any truly good developer with even a basic interest in OS design should be able to understand and implement the basics of kernel level pre-emptive multitasking and cooperative threading in year one of college, but these geniuses still have you studying the basics of semaphores and mutexes in your senior year.
Let's look at some of the Cornell senior year projects:
http://people.ece.cornell.edu/land/courses/eceprojectsland/S...
Seriously? Sound identification of Loons? Arduino-based ANYTHING??? Touch screen user interface for a car using LabView?
These are the "geniuses" todays top schools are turning out.
It may sound like I'm calling out these specific people for their lack of imagination, but the fact is that they are the majority, and some of them go back into the educational system to get their masters in education so they can teach! Sheer insanity.
I've worked with hundreds of developers and mentored a dozen or so; some excellent, some not so much.
The one thing I can say without exception is that I have never met a truly prolific architect level developer that wasn't self taught. Most of them had a passion for programming, and an education in some other speciality unrelated to development.
Yeah, you can get some basics, and yeah, there's SOME value in compiler design, but nothing you can't get from "Dummies Guide to Compiler Design", which, if it exists, was probably written by a student admin ghosting for a tenured professor.
What a system.