I gotta admit, I do this sometimes. I’ve also recommended avoiding it, but it isn’t unusual for me to have an Assembly that’s well over 100’ long, and mating back to a component at the other end sometimes just isn’t practical.
I do believe I have fewer of those mate errors that go away with a rebuild when doing this than I did 8 - 10 years ago.
I really hate it when I get to the bottom of a tree and there are a bunch of Delete Face and Move Face features. Somebody usually inherited a design, and couldn’t be bothered to go back and figure out the feature tree, so they just hacked and slashed.
So when I inherited a bunch of parts like this, I just remodeled all of them. It was easier than figuring out that tangled mess.
There was one time in my life when I used Move Face. That was because I had a linear pattern of SM parts and the ones at either end (one was the seed for the pattern) needed to be from 1/2" material instead of 3/8". BUT…I made it parametric and linked the Sheet Metal Thickness property to the actual dimension for those two.
My main pet peeve is hearing so-called power users stating to their managers that: “I am slow because SOLIDWORKS is slow. There is nothing I can do, we have to accept that the software is bad.”
There is also a lot of satisfaction to be felt hearing the retraction after they learn the proper technique to get things done.
Here’s one that annoys me. Just the opposite of what someone else posted. When someone picks 100 edges in a single fillet feature, then the fillet on one edge fails or loses its reference, so the whole feature and all edges fail. Instead, if you have several smaller features with fewer edges selected, troubleshooting is much easier. Plus, if you just select a single edge instead of a whole tangent group, the software can pick the rest of the tangency automatically.
It wasn’t a co-worker, but I received a drawing from a client once that specified the embedment depth of an anchor bolt in concrete down to the 1/16". I’m guessing whoever did that has never seen anyone setting forms, tying rebar, or pouring concrete, and has no more than a vague idea, if that, of how the process works.
By the way, in case anyone is wondering, there was absolutely no need for that kind of tolerance. I’m not an engineer, but I don’t need to be to know that. Specifying embedment depth to the nearest 1/4" would have been just fine.
I’ve also seen rebar bends dimensioned to the 1/16". I will occasionally dimension rebar to the nearest 1/8", but I cringe when I do it. That kind of tolerance in concrete work is rarely, if ever, needed, and in any case is just about impossible to accomplish.