Karmashark raised a good point about collateral requirements.
Lets open up a discussion on masternode collateral. Here are some options:
We do need to seek views as people will have opinions on affordability of masternodes and distribution around early investors. We will have some limits as to what we can do to please everyone, but lets give it a go.
From my perspective, 10,000 HLM generates too few servers. We'll be looking to run millions of user accounts so we have to take into account the numbers of severs we can run on the network to spread and carry the load.
We also need to take into account 5%-20% of nodes going down at any one point due to DDoS attacks; day to day redundancy; and network performance as experienced by users.
250 HLM is towards the range we're interested in, but we won't need that level of capacity for a decade or more, so it creates a sybil risk until the price of HLM rises.
800 HLM to 1,250 HLM for collateral is the range that gives a good compromise between economic barrier to sybil attacks and providing sufficient room for supporting ordinary payments users on the network.
Then we need to factor the current and near future coin supply. This has an impact on the choices we make now and in the near future.
We can change the rules down the road, but that would disrupt business as usual for everyone running masternodes. Running a 24/7 payments network for tens of millions of users means we'll have to look at migrate+upgrade (really hard) vs upgrade only if we ever need to switch the protocol while trying to meet end user SLAs.
We will have 2 or 3 hard forks between launch and the end of 2018. We won't be anywhere near capacity by the end of 2018, so we can come back to this issue in 2018 when we start planning for something that will last for 10-20 years in terms of the collateral vs. server capacity.
