Hi Chris,
That was an interesting blog post, if a little depressing in places.  Thanks for 
that.  I've been rather out of touch with JISC developer activities for the past 
couple of years or so, having been sucked in to an EU project, and currently 
planning a "sabbatical" with at best part-time employment for a year or so.
I've been fortunate enough that I'm in a position to pursue a slightly different 
model.  Roughly, I'm putting a few months of my own effort into an open source 
project that I think may fill a gap in the RDM landscape, and also scratching a 
few personal itches ([1], in case you're interested).  If I'm right, I'll end up 
with the bare bones and proven basics of a tool that can be adapted relatively 
cheaply to meet specific researchers needs.
Your comments about no more small projects were disappointing to hear, as I had 
been hoping to tap small-project funding to support the specific adaptations, 
while keeping the core system more of a generic tool not specifically tailored 
to HE/research.  My intent was that it /should/ look like very good value for 
money to a funder like Jisc, as they wouldn't be paying for the experimental 
foundational work, but just the "superstructure" elements that deliver concrete 
results.  From your recent discussions, does that seem like a non-starter with 
the "new Jisc"?
Regards,
#g
--
[1] 
https://github.com/gklyne/annalist/tree/develop
     - it's still very work-in-progress,
       aiming for a demonstrable prototype in the next month.