There is a whole topic discussing this, maybe you could look at it. I am not sure of the link at this time and I cannot find it either.
One of the big issues was the buffer that some 80 or 85% of tokens available at the start allows the network to be a lot more dynamic, allows it to cope with so much more when events happen. For instance a unusually large inflow of data, or a unusual outflow of rewards.
I am of the opinion after that discussion that distributing all the tokens at the start is the wrong way to go.
Not against there being a higher total to start with (eg 50% given out) and MAID people receive a proportional amount of SNT for each MAID.
Cannot see any real issues with say 50% being given out for MAID (old 10%) + the (old) 5% reserve. There would still be enough of a buffer.
But the 4.3 bn supply under the current system is a myth. The recycling of tokens means the supply to be sold over time can be 50, 60, 70, whatever bn. What will the market actually see. Perhaps it will only see the amount estimated to be “out there”. If its estimated there is 2 bn in the hands of holders then the market sees 2bn, and really would this not be the amount they should be seeing.
And would not having a pool at the start be wise? Well the original plans was for the pool to be that 80/85% of 4bn. People went into the ICO believing this was the case. As said above maybe 50% might be a good compromise if you feel there is a need to change the original concept.
Also what is the actual benefit? Will people see their SNT worth more? 6.7SNT with 4.3bn available worth more than 1SNT with 0.64bn available. More than likely the same value. So why change what people bought into the ICO with and believed for all these years to be the case? There needs to be a very good reason for doing so in my opinion.
The buffer in the network cannot be underestimated and you even suggested a work around for this. Why have a work around when the original did it for you anyhow. See below comment.