Although I agree with his assessment of the memory contraint in XEL making it useful to only limited use cases, I really don't understand the concept of integrating BOINC into a coin.  First, we already have GridCoin doing a pretty good job of that...but if you look at BOINC as a whole, I don't see any need a cryptocoin to draw people in...look at how much interest BOINC already has.  Also, I've never liked the idea of POW for any of these science coins...I don't think it helps, but it does complicate things.  I personally would have preferred if XEL got rid of it.

I always figured the market for XEL was for research on problems that aren't necessarily large / glamorous....maybe just some simple research a scientist needs extra computing power for, and is willing to pay a small bounty for miners who solve the problem...but I do agree XEL servers need to suppor larger memory usage...just not sure if it's possible in this current design.