currently there are 10 formats, and already it happens sometimes where two players are online, they both want to play different format, if we introduce another 10 formats (with erratas), then this will happen even more often, that players will be willing to play different formats
If this is a problem, it means that before we start work leading towards doubling formats, we should work towards doubling (tripling etc.) number of players. There are many things that can be done about this and majority of them does not require coding skills at all, so it's a good thing for volunteers to do (I would gladly help in this).
Old formats must stay available. At least for casual games. And an "oldtimer" tournament from time to time would not be bad either.
Before the game grows big enough to support doubling the number of formats, I think we should do no changes to cards. Looking at the votes in the poll, there is way too much to be lost if we force changes on everyone. I voted yes, as I would like the game to live and evolve, but I see that there are many people who enjoy the game as it is and don't want it to change.
I highly doubt that there is a single person actively playing Gemp who decided to try it out based on the assumption that it was going to be a brand new experience--we all joined because we wanted to play the game we were nostalgic for.
I know of at least one player who has never played LotR outside Gemp. And I think the first thing to do is to find ways to attract such players to the site. Every game needs new players to go on, as old ones leave from time to time. And there are many people we can aim at - interested both in gaming and LotR story, but never involved in LotR TCG during its days of physical cards playing due to one of many possible reasons:
- being too young at that time (FotR was released more than 10 years ago, Age's End nearly 5 years ago, a lot of people have grown up since then),
- not wanting to invest money into a TCG - Gemp does not require this,
- living in a place where there was no active playing group.
if you want to have 2 card versions officially supported in 2 separate mirrored formats, then when you open a booster of a set, you'd have to have 50/50 chance to open the new or old version, that means that user will have 50% chance of opening a version he doesn't want/need
Every time you open a booster, you are anyway more or less sure of getting at least a few cards you don't need

But, for this problem, a fix is easy. Whichever you prefer: either make the card count for both new and old version towards collector's tournaments (those are only where this matters) or mirror the boosters as well - for each set having cards with old and new versions, there will be a possibility of buying a booster which contains old versions and one containing new versions.
Also, if I did that, and had one version of card, then once again we would be having a "hidden" card text, where errata is not on the card in game. Not when you open the booster (how do you know if you should show old or new version), not during deck building (which one to show), neither when you actually play the game (how the JavaScript client should know which one to display?).
When you open the booster: doesn't matter for the general view (cards are too small to read), both versions in the magnified card view (added bonus: players get aware that this particular card has two versions). The same in merchant and deck builder. During the game: each table has a format, each card has only one version for any given format, I guess that's enough information to learn the program to display the correct one.