Regarding that nasty “BKS distribution” stuff and just to give you an impression about the numbers:
say exactly 50.01% have updated their client and would vote for a BKS grant as well to distribute those BKS to 4.0 minters to increase the ratio of updated clients (precisely: clients minting with 90% of minting BKS) to above 90%.
That’d mean the total amount of BKS would roughly be increased by factor 41.
Without taking psychological effects into account (and just trying to deal with numbers), that’d reduce the value per BKS by factor 4.
The BKS would be distributed to BKS addresses that signed protocol 4.0 blocks in an announced time frame which needs to be big enough to give small holders a chance to mint.
Whether that is perceived as ruthless action and harms the BKS rate even more than by factor 4 or is considered a measure that shows how much BKS holders want the exchange to succeed no one can foresee.
An announcement of this measure alone could wake up BKS holders.
Even if it were taken, it could be done in more than just one round.
There’s a lot that needs to be defined before.
Distributing the BKS would be a pain.
I don’t know whether that can be done from multisig addresses.
I suppose we discuss the details once it’s become clear that such a measure can’t be avoided.
I just wanted to explain why not all is lost 
1
for the sake of simplicity say 100 BKS exist.
50 are on protocol 2.0, 50 are on protocol 4.0.
To bring protocol 4.0 at 90%, you need 400 BKS and distribute them to those addresses that vote for 4.0.
A grant to create 400 BKS passes.
After the 400 BKS are distributed, 450 BKS are on 4.0, 50 BKS are on 2.0.
450/(50+450)=0.9=90%
Mission accomplished.