And MIDI, doods... help out kylo if you can. I'm out of ideas.
 

It's a good theory, but in some cases the 2 clips have never been linked together at all. They were produced by dragging from the same plugin but that's as far as it goes.

ampfixer
I've had trouble with clip management ever since comping was introduced. I've seen the issues the OP has with locks toggling in clips and the weirdness of clips acting locked when they are not. I also don't use comping or take lanes but according to Beepster's post, we are all using take lanes even if we don't want to. It bothers me enough that I've given up using midi patterns or step sequencer to assemble midi drum tracks. Clips have developed a mind of their own.

Kind of yes and kind of no.
 
To me it's a culmination of...
 
a) the Smart Tool... which is ultra crammed and insanely context/area sensitive with seemingly endless and ever changing tools. So much so it itself seems to get confused as to which form it should be taking on at any given point in time. Without knowing exactly what the tooltip should look like, what it should be doing and where it should be doing it it's almost impossible to use. I've kind of got the hang of it all but still often times I revert to the older, more dedicated single purpose tools just to cut through the logistical nightmare (and sometimes outright refusal of the ST to switch to what I what).
 
b) the way comping has drastically altered the old school way of doing things. Comping is the new way and although the old workflows CAN be acheived you still have to really know comping through and through to completely revert back to those old school methods.
 
All in all, to me, after learning how it all works I do think it's a great workflow but once X3 came out (the first with the fully implemented comping features) it seriously disrupted my editing until I did actually sit down for a half hour to specifically figure out WTF was going on.
 
Now I fly with it but I also have to keep all this stuff in mind from the time I hit record until I flatten my comps (and in many cases drag the comps into a clean track).
 
The absolute BEST thing to do for those not used to it is crack open lanes right from the start of tracking to watch them get inserted into the lanes and keep them open (or crack them open again) for editing... and edit in the lanes.
 
There are some extremely helpful vids in the CakeTV area that explain (mostly) what's going on with comping.
 
Aside from that... yeah, we're kind of fuxxored for just straight old school track editing. All that is a massive conceptual nightmare in all DAWs since layered style tracking has been introduced anyway though. Always something weird is gonna happen or have to be forced. Seems Cake have just said "f*ck it... it's gotta be done in lanes".
 
I am (mostly) okay with that and have my little workflows to enable old school track editing.
 
I DO most definitely want to be able to archive and hide lanes though. That would make things an arseload easier.
 
Cheers.

Kylotan
It's a good theory, but in some cases the 2 clips have never been linked together at all. They were produced by dragging from the same plugin but that's as far as it goes.

No offense, but frankly it makes more sense to conclude that you're mistaken about how the project got to this state than that a bug managed to do something as complex as linking the properties of two instances of an object imported from outside SONAR's project space as separate actions.
 
There's a bug alright, but I think it's your typical "dumb bug of omission", not some super-smart bug intent on locking all your data against your will. Bwahahahahaha! Your base are belong to us.

icontakt
As for the selection issue in post #1, I can't tell until you post a video in which you zoom in horizontally (by keeping hitting Ctrl + right arrow, for example, until you can no longer zoom) at the start of measure 331. Clip Properties Inspector doesn't always show correct information.

Good catch. Looks like all my clips were aligned correctly, but somehow the marker is 2 milliseconds after the measure. I have no idea how that happened.