This is another reason why I feel like trying to work with existing installs as anything more than a source to copy VP files from, is going to make this project complex to the point of likely failure.
I'll reiterate that, I think that for the success of this launcher/installer, it needs to work with an ecosystem built from the ground up. Any support for existing installs needs to be minimal, like I said, maybe we scan for VP files we need to download and copy them into the new structure if necessary, but beyond that, the added complexity seems to me to outweigh any benefits.
The MediaVPs are not really an exception here, I don't think. They need to be manually converted into a Knossos style dependency format just like any other mod. But as there are different major revisions of the mediaVPs, I think we might need a concept of slots. Or at least, as stated, we don't treat 3.6.12 and 2014 as the same mod with different versions, but yeah, different mods. They could both be selected as primary mods, or individually be a dependency of other mods.
We will probably run into the need for this with other mods as well, so I am hoping there is a framework for this that can be applied as necessary. If people want to play "Media VPs FS2" they probably shouldn't be provided with multiple versions by default, just the latest. But there would be cases where someone might need to launch an older 'slot' of a mod, maybe someone wants to launch Diaspora R1 after R2 is released, an old MediaVPs version, etc. That might be better kept under the developer section or something. But to keep the initial interface clean, we need to ensure we don't overflow it with deprecated slots of mods.
I agree that 2014 is a bad version number, which is why I think each mod needs to be brought into Knossos with entirely new versioning where possible. We may not even be able to go that far back in time to find mods that currently work in the new ecosystem without some work cleaning them up, so reusing old VP files, and worrying about old version numbers is something I think we can avoid.
My main hope is that this new system looks so promising that it reinvigorates the Campaign Restoration Project and other people to start updating mods to the point of at least getting them running on recent builds so they can be added to Knossos and made available. It would be a significant undertaking perhaps, but that's what the CRP was created for. Karajorma has been working on TBP for instance, and may be interested in using Knossos for its next release. I would still very much love for this to be the platform for FotG when it is publicly released (hopefully very soon), and I keep hearing that Diaspora is approaching R2 (seen the new eyecandy shots?). There's three different total conversions that could very possible be on the coattails of the 3.8 release. If all of this can fall together it would be a huge incentive to get FS2 mods updated to working with recent FSO releases for inclusion in Knossos.