Fact is, even with VP being a not-out-of-the-box format, people still modify it. We still have user made tools for modifying them. And people STILL break things when they do so when they don't fully understand what they are doing.
These things will continue, regardless. OUR job is NOT in making that more difficult. Especially not for an "OPEN" code project that is driven by Community participation. The end results of a mod team (Or the MediaVPs) is that it should just WORK once completed. And the more shenanigans you put into place to prevent users fiddling with it, the more likely your are to introduce something that breaks that basic capability.
As The E has rightly said, we can checksum Official packages very easily (via Installer, engine, Multi, etc). That the tools far creating/modifying becomes a step towards better accessibility to being able to mod and distribute is a positive we've been waiting for and cannot pass up on.
Making it easier to be able to be stupid also means it makes it easier to do it right. What impacts one will always impact the other, so we need to focus on the benefits obtained by making it easier for everybody and let time and education take care of the mistakes.
That and we cannot deprecate .VP. That's against the essential tenant that we have to maintain RETAIL compatibility. But it opens the door for being able to get around other issues. We can expand file name handling with the new format to being greater than 32 char's without causing a potential negative impact on the basic Retail behaviour. (Especially since if we tried expanding that limitation WITHIN the VP format, we BREAK all the existing VP tools out there). This is less a concern if we just attribute the new format to handling that.
Further, with the new format, we can have the creation of .manifest files (in addition to the checksum process) to validate not only the package as a whole, but all the bit's in it. And being able to have (eventually) the Installer/Launcher/Whatever be able to merge Patching Data into Official archives (when from an Official Source) so that each new release doesn't mean an complete re-download of everything to stay current. This system would be much easier to implement with this new format without introducing potential complications in being able to load currently existing mods or Retail data. (Mind you, this is merely a -potential- usage scenario, not an immediate use qualification of the new format).
Anyway, running out of steam on this one, so I'll stop while I'm ahead.