I use an xmp based workflow, but I'm now doing deflickering in post with
this script I wrote, rather than in camera with ML. It uses basically the same algorithm as ML post deflicker, but has a little more flexibility since you're doing it in post (you can run it again with different settings, don't have to worry about missed frames b/c your interval was too short, etc.)
I'm using xmp, but even if alex is the only one with ufraw I'm in favor of defaulting to open source software in ML (a similar thread here: http://www.magiclantern.fm/forum/index.php?topic=9078.msg85754;topicseen#msg85754)
If you want to get technical, even though XMP is going to be typically utilized by users of comercial software (ACR/Lr) and UFR used by those using open source, XMP itself is an open standard (ISO 16684-1). IDK if the same can be said of the UFR format. Adobe has also released the code for their XMP Toolkit under a BSD license.
That being said, I don't really care which is the default, I don't think it's a big deal, and this should be removed from ML anyway once post tools to replace it are complete (I've tried to do my part on that!), since this is something that should be done in post not in camera.
