[TheOtherCliff]

ManualCommandReadOnly() do not set something, this is a check function returning the uavo status

Yes.  My point was that he needs to look at this code.

[TheOtherCliff]

That could be a possibillity, but why do I receive a changed value when I send and receive the package again, if the other source overrides it?

I am making a guess, but normal input runs maybe 50 times a second.  You set, then 0.1ms later you get, and the value is what you set.  10ms later the normal input runs and overwrites your set.  You could wait 50ms (really only 20ms is probably needed) or so between set and get to see if it has changed back.

Is there no other way?

You don't have to use the LP framework, but you do need to do the things that it does.  In particular, you need to set that "readonly" property to turn off the normal input processing.  I suggest you look at the C code referred to look at what is exectuted when the readonly property is set. 

[f5soh]

I suggest you look at the C code referred to look at what is exectuted when the readonly property is set. 

Not sure you are able to play with metadata from arduino, without additional code.

Is there no other way?

As a test, you can flash the firmware attached (16.09) without receiver module.
You should also consider using Next with matching arduino templates if you plan to use others UAVOs.
Arduino target was added after the 16.09 release, not sure all the UAVO Object IDs defined in firmware are the same as the ones used in arduino templates you are using.

[M3troiX]

You should also consider using Next with matching arduino templates if you plan to use others UAVOs.

I've tried to build the GCS from the "next" branch, but I there is an error with the coptercontrol.elf. Enclosed you can find a screenshot of the error.

[f5soh]

Seems a missing python2 package