(Not from management discussion, just my input on this)
The problem with this is that to do this effectively, we’d have to make the limit very small, and at that point it wouldn’t even be worth having. Just as an example, say we limited it to 10,000 blocks as shrimp suggested. A 101 blocks sided square, or a 22 blocks sided cube would be more blocks than the limit. It becomes too small to effectively work with, so yes, you’d get access to worldedit commands, but you don’t get most of the functionality you were actually asking for.
Any kind of test we can come up with won’t be able to determine how well someone actually knows how to use worldedit. Sure, we can check that they know how to set and replace blocks or whatever, but that isn’t enough to make sure people, for example, know how to see if their selection is wrong before they enter a command.
And don’t even get me started on the brush tools. They are extremely difficult to use properly, and you can very easily and quickly make huge amounts of edits, to the point where you can’t even undo it any more (worldedit has a cap to the number of edits it saves in memory that can be undone using //undo) when you realise you made a mistake.
Even if we found effective solutions for all that, the server we currently have is not capable of having 10+ people all using worldedit at the same time. Plus, all these worldedits would need to be logged, adding even more strain. We hardly come close to reaching our donation goal for the current server, we’d need more to handle all this worldedit. As spec once said, we can either be an 80 slot server where most people can’t use worldedit, or an 8 slot server where most people can.
tl;dr:
In order to prevent accidents we’d have to limit w/e so much it’s not worth having.
Tests won’t prevent accidents either.
Brush tools cause even more problems.
Even if we got over the rest of the issues, the current server can’t handle it, and we don’t currently have the money for a suitable upgrade.