Recently, there was a discussion among faculty members and industry folks surrounding the role of training PhD students suitable for the workplace. The synopsis was something like this. Many students who get PhDs are interested in staying in academia but the realities are that for every 10-20 PhD students graduated, maybe 1 of them will get a faculty position. So what are the other PhD students supposed to do? Well, get jobs in industry. Except most of them have been trained to have an academic perspective on things and have not been trained to understand effective project management, deadlines, deliverables, etc.
One of the ideas to solving this problem involves having the PhD student propose their thesis topic earlier in their training. The idea is that if the student identifies the scope of their PhD earlier, they can then focus more on getting those "deliverables" done. Also, from an advisor perspective, this is a contract of sorts where the PhD student can keep the advisor from moving the goal posts because there is an established stopping point to the student's work.
While this is a grand idea and I whole-heartedly agree with it, it is simply not practical. And there is one simple reason for it. Money.
The most direct way that I see this relates to the size and duration of research funding awards. In my area, most research awards are 2-3 years with insufficient funding levels ($100k/yr) to establish instrumentation and equipment to complete a task. Thus, a student may start out on one project but because of my funding situation (for which the student has no control), that student may be forced to switch to multiple projects just so I can pay their stipend and satisfy multiple program objectives. However, these projects usually are not interlinked enough to draw a common thread for writing a thesis. This makes defining thesis objectives very nebulous when you don't know what task you will need to do to satisfy which program manager 6 months from now. I spend more of my time (probably 60%+) going after money than I do doing anything else, including actual research, teaching, service, mentoring, etc. As the PI, frankly I'm trying to solve the paycheck problem for my students first and then worry about the details like defining a thesis second. It is a sad state of affairs and I am disappointed in this but that's the practical realities that I face.
The second way in which I see an issue with training PhD students for jobs in industry is that no funding agency actually awards funding for practical, applied research. (At least funding that is not ITAR controlled.) Virtually every response I get from review panels is "Wow, so-and-so company that wrote you a support letter should directly fund this.". I'm basically convinced at this point if you send anything practical to NSF it will get shot down for not being sciency enough. But they neglect that if they fund that research, it may turn into technology to spin off from the university, which may create jobs, which grows the tax base, and increases funding for entities like NSF. (BTW, that's the NSF standard lip service as to their motivation for funding projects).
I've had a $2.1M NIH BioTech R01 turned down last week, an $875k fellowship turned down today, and two NSFs for about $800k turned down last month. All of them were doing some amount of science but all of them had practical applications which would be excellent training for students.That leaves 5 of my students with cloudy funding futures, several of which will have to put off their thesis proposals, because I can't secure funding. And that's not their fault. But the really infuriating thing about all of this is I have no shortage of companies that explicitly ask me to notify them when I have students graduating because they want students who have been trained in my area. But I fear I won't be in academia long enough for that to matter because the practical realities are that I need to bring in enough money to support the students and the funding agencies simply do not fund areas where students may be practically trained.