@bbcreporter: it is about scalability and preventing blockchain bloat. outsourcing all the beautiful functions of nxt that are not relevant for network security to childchains allows to prune the history of these much used smart transactions. the mainchain will only keep snapshots of each account's holdings of the childchaintoken, not the complete transaction history of these tokens. this leads to less blockchain bloat. less bloat also means that more transactions per second will be possible, because you can use the saved memory space to increase blocksize.
Now I understand clearly. This is an elegant solution if you do not want to start over and lose the original NXT. This also good as a sort of revival if you want user interest to come back.
Thank you for your reply.
No. It's the grandma of all elegant solutions

By distributing 100% of the security relevant ARDR tokens to all existing NXT accounts you get the best decentralization and therefore best security for the new platform. Still NXT won't get worthless after the snapshot phase, because there will be another snapshot of NXT holdings in mid 2017 which determines the distribution of the first childchaintoken. If the team doesnt get funding before, they will ICO half of that token. Dev funding problem solved. And even after that Nxt 1.x will still be fully functional, supported and therefore not go to zero.
There is elegance not only in the tech, not only in the management of the transitioning from Nxt to Ardor which gives desperate whales a chance to cash out and makes hodling investors happy, but also in the marketing possibilites that come along with that. This is the result of arduous discussions within the Nxt community. Despite all the naysayers they found the optimal solution and squared the circle. What a great team

Reports about Nxt/Ardor should run on bbc news all day
