Call me a pessimist, but I just can't see it ever happening. Getting stuff like this done either requires you to be close with an OEM or make it up entirely from scratch, which results in $350+ price range, which most people are unwilling to even consider as the past has shown...
"I hope the miniguru kit will be under 100 euro
"... yeah right.
People are thinking too unrealistically for this to work, I think. Another problem is the need for equipment and knowledge. The OTD guys took several revisions for their custom boards to actually be decent. Prototyping such a project is a painful task, especially when each re-try and re-adjustment again costs you so much in tooling costs and machine leasing fees, etc...
To realistically get a project going at least the following things are needed:
1) software/firmware programmer
2) hardware guy(s) who work out a good matrix and custom controller.
3) If you want a small size keyboard, you will need the controller on board, since a Teensy or such will not fit in the case. That means you will need a PCB design guy that knows his way around micro controllers, as well.
4) If you want a metal case you need someone who REALLY knows his way around CAD design and the machines that the files are being used on. With the prices in this business you can not afford many prototypes and will have to get it right in two or three tries at most
5) Assembly - Once you got the PCB and case design done, who is gonna solder in all the parts? SMD soldering is serious business and I think most users would be afraid of it. I know for one that I am. So say you have managed to get a setup for 50 boards going, you will need someone to solder in all the parts and assemble the boards accordingly
6) Parts - Custom keycaps are somewhat easily thanks to SP, but will probably come in at about $50+ per set alone with an order of 50~100 sets. Besides that, you will need small but important parts like stabilizers. These will need to be ordered in large quantities from Cherry directly, or taken from donor boards. Option A is costy, option B pretty unrealistic
These are just some of the things to actually consider, maybe it helps in the planning. If you do indeed come up with something that gets past rough ideas, I'll gladly take a board.