I would go as far as to say that we need generalists who will take the time to chronicle and revive works from the past to make them relevant again. Better sti

I would go as far as to say that we need generalists who will take the time to chronicle and revive works from the past to make them relevant again. Better still, we need the mathematics community at large to become more accustomed to seeing papers published in journals that do this so that a larger contingent can be made aware that these papers/ideas/theorems/corollaries/lemmata/definitions/people/pictures/etc. exist. Note also that this doesn't necessarily preclude anyone in mathematics from adopting a specialization. In my eyes, the younger mathematician is in the same place as the startup founder: in the position to take a big risk and do something that sounds like a stupid idea (e.g., avoid publishing results anywhere but freely and openly on the WWW; see also: Daniel Bernstein) in order to have it pay off (e.g., earn tenure).