I'm not sure whether this question is relevant at the time of writing. IMHO, PhD is a highly institutionalised agenda and why haven't we seen coaching for that matter done online especially where bioinformatics is concerned? For instance, I would really be interested to have a committed online mentor if I were to take PhD up.
What sort of projects will be relevant if this kind of system is permissible?
Another argument against is that it is very difficult to monitor the progress that a student is making without face-to-face meetings. Are they actually doing any work? Are they working on appropriate problems? Are they having difficulties? Do they fully understand what they are doing? All of these things are much easier to ascertain in person.
Effective two-way communication is very important in education, research and any kind of job and it is not very practical online.
A PhD is, at least in my eyes, a traineeship. In the long term, it's not only the result that counts (the what), but also how one got there, what has been learned, how much interaction has been achieved... In that sense, it is a tow-way communication indeed... or even more than two, as underlined by niallhaslam, that would completely missed by working in isolation. Science is a social business!
The reason why PhD educations follow an "institutionalized agenda" is that PhDs are awarded by institutions, typically universities. Unsurprisingly, institutions have their own agendas. A very important point on any university agenda is teaching, which would take up way too much of the professors time if part of the work load could not be delegated to PhD students. This is one reason why universities would be hesitant (at best) towards the idea of online mentoring of PhD students.
Thought I would throw in this link - not because I particularly believe it though he is a good writer who not infrequently predicts major trends - as it may shed some light on the arguments in favour of remote access and supervision. He is making the argument that schools, i.e. high schools, will be irrelavant in the future.
Cringley, X
For those that don't like reading links, he basically argues that with the increasing availability of information, e.g. OpenCourseWare from MIT, that the boundaries that once separated and required educational institutions are now starting to break down.
I personally doubt this will ever really happen to PhDs since there is too much to be gained from having high concentrations of students together in one town be it Boston, Cambridge (UK), Heidelberg, or whereever. A lot of what makes the best PhDs (to me anyway) is that they have access to opposing or at least differing perspectives in the form of other students from other disciplines. Probably doesn't seem very relevant in biology at the moment where there is the tendency for collaborative work and consensus building. But in the future it will probably be useful to again have people that are prepared and trained in standing on their own two feet and holding on to an idea despite the opposition. Meeting other students in the pub after work (or other cultural equivalents) and arguing for one's idea is the only way to achieve this (ok not only but its a great way). Whilst the organisation of a PhD is a formal process most of the training comes informally.
On the other hand with many bosses having at least two affiliations - to what extent is this happening already? You may in some labs get a meeting once every two months and the rest of the discussion is over e-mail and skype?
A PhD is, at least in my eyes, a traineeship. In the long term, it's not only the result that counts (the what), but also how one got there, what has been learned, how much interaction has been achieved... In that sense, it is a tow-way communication indeed... or even more than two, as underlined by niallhaslam, that would completely missed by working in isolation. Science is a social business!