Interesting example @Dimitar from the Omni(Mastercoin) fork Counterparty. Also illustrative of the amount of exchanges and liquidity levels available to ERC-20 markets for tokens.
Two way conversion sounds like unnecessary complexity right now. The script to convert X to Safecoin would just need to accept the amount to be converted, and that amount could come equally easily from the Omni burn or SafeCoin(ERC-20) burn script. I do agree that later on a two way conversion would be useful but between SafeCoin(ERC-20) and Safecoin to create a bridge between two active communities allowing investors and developers to cross and build cross chain services with minimal friction.
This is a fairly strong statement to make right out the gate. If we take just the community poll numbers at face value the majority of just active forum readers appear to be expressing interest in participating in Safecoin(ERC-20) markets. How much that translates to actual demand (from just existing forum members) is hard to quantity. Some of us are unable/unwilling to participate in existing two Omni exchanges for the red flags raised previously and in the very limited cases I personally know of it is not an insignificant amount of monthly income streams/xmas bonuses that will be now cut off from MAID. We will have a better idea of the bigger picture in a few months I am sure.
Complexity wise there are a few older mostly pre-Ethereum projects that transitioned to ERC-20 token (many more have jumped from ERC-20 to Tron and EOS) without apparent user confusion so I am not sure why you think it would be different in our case? Your initiative to study similar transitions like @Dimitar StorJ example is really great. It might also be very useful to quantify the range of exchange service options available to the similar caliber projects at least from an active development yardstick. Here is a list of potential ERC-20 projects ranked by development every month. Just eyeballing the number of exchanges available to trade over on coinmarketcap I would guesstimate an average of 5 exchanges for actively developed projects. I may have a biased opinion but the Safe Network project has way more potential and is of a higher caliber than most of those projects and yet is seriously lagging on the exchange front. The Omni effect.