Getting PhD and trained as scientist has a HUGE value counter to the arguments in this article. First, as Guy Kawasaki puts it, you want to be in school as lon

Getting PhD and trained as scientist has a HUGE value counter to the arguments in this article. First, as Guy Kawasaki puts it, you want to be in school as long as possible. Getting a job means life long working on your bosses orders whether you like it or not. Going to graduate school allows you to work on your favorite area, become master at something and gives you an opportunity to add that little bump in the human knowledge with your name. After you get PhD, you should also realize that not everyone is cut out becoming next Einstein or Feynman. If you can get good postdoc with interesting challenges then fine but otherwise you can always go to industry and get a job. You enjoyed your years in graduate school unlike spending those years trying to make some company little bit more profit and that's what counts. Today, having a PhD in any field is invaluable if you want to enter the software industry. Most PhD work today involves some level of data science and software work anyway. When you apply for a job, at least in software industry, you will get better offer faster. The correct way to look at this not how much money you would have made if you weren't doing PhD and straight went to industry but rather how much fun you would have missed out on your favorite area of work and possibly childhood passion if you didn't. Software is eating academia. Concern for not getting tenure positions should be minimal today. There has never been better time to get PhD and get trained as scientist.