I don't have much experience with jsgme installs ... I usually do manual installs in my #DBW folder ...
Now I perfectly understand the concept of making different air.ini files for different theaters of war or periods. It seems a very simple and sensible way of working around the java limit. But a question: I usually leave the mod files themselves in #DBW, even when I disable an air.ini entry. Sometimes the modders give the most unusual names to their folders and it is difficult to track them down again. I know they can be renamed, but that can also give probs with updates and all.
So I just leave the mod folders in #DBW, not disabling them, this makes it fast and easy to enable/disable air.ini entries. . Does this affect the java limits in any way? I read that a disabled air.ini/stationary.ini entry will not read the corresponding javafiles, but is this certainly so? I would just like to check with members who know about these things.
I can answer you by experience: It can affect them, and sometimes I had to disactivate a mod completely to make others run smoothly. By now, I had only three cases: Fiat G.56, Swordfish Mk.II and lately the Bredas Ba.65. So yes it can affect them, but these cases had been rare. Most of them, being there enabled in the JSGME but not loaded from the air.ini, there's no visible difference.
That's why I am talking about a proccess of standarisation, to ensure that all of them together will work smoothly in every installation.This discussion has already taken place, over the course of several months. Everyone from the modders, to mission and campaign builders, to the regular user was allowed and even encouraged to participate. Many options, including the not-so-revolutionary method that you describe were considered and the decision was made to go with the simplest route.
Why reopen discussion on a ,matter that has already been discussed ad nauseam and come to a conclusion?
EDIT: Just checked my PM's. The public discussion about slot merges yay or nay, and other methods (including yours) go back to mid august!
Day late and a dollar short I would say.
I don't claim that I discovered the wheel. My contributions to IL-2 development have been very poor compared to others. The point is that I tested, it worked. I have now in my #DBW folder, 3.85 GB, 33.312 files and 1.115 folders. 57 mods activated (some of them merged mods that are available normally only as seperate installs).
My wrapper says
Scanning #DBW folder took 223 milliseconds.
Total number of modded files = 32961.
Sorting modded files list took 8,080 milliseconds.
Removing 84 Duplicates took 0,661 milliseconds.
Total files opened = 81561
Total search time consumed = 4,162 milliseconds (0,004162191413 Seconds)
Search Time per File = 51,032 nanoseconds (0,000000051032 Seconds)
Average Search Iterations required per File = 8,5
So the difference is that there is no theoretical discussion, it's tried and tested. I pushed it to its limits. That's the only substancial thing I did, but hey, I did it.